Method and device for automatically arranging business relation chain and storage medium

Abstract

The invention relates to the technical field of automatic arrangement, and discloses a method and device for automatically arranging a business relation chain and a storage medium, and the method comprises the steps: constructing a business rule base based on supplier information, customer information and business information, and carrying out the standardization processing of customer demand data, and obtaining the standardized business demand information; performing multi-dimensional matching operation on the standardized business demand information based on the business rule base to obtain an initial supplier set, and performing dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence; constructing a chain node graph according to the optimal supplier sequence, calculating and verifying a business constraint relationship between nodes, and generating a business relationship chain; the business relation chain is converted into a standard order format, order data synchronization is completed and an order synchronization result is output through encryption signature and a concurrent distribution mechanism, and the accuracy and efficiency of business relation chain arrangement are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic scheduling, and particularly relates to a method, device and storage medium for automatically scheduling business relationship chains. Background Art

[0002] Traditional business relationship chain scheduling mainly relies on manual operations, which not only involves a large amount of work, but also is prone to data errors and judgment deviations, making it difficult to meet the needs of modern enterprises to quickly respond to market changes.

[0003] In actual business scenarios, the business relationship between suppliers and customers is affected by various factors, including product quality, delivery time, price range, cooperation history, etc. There are complex interactions and restrictive relationships among these factors. Manual scheduling methods are difficult to comprehensively consider these factors, often leading to problems such as inappropriate supplier selection, unreasonable resource allocation, and low business efficiency. Summary of the Invention

[0004] The present invention provides a method, device and storage medium for automatically scheduling business relationship chains, and the present invention improves the accuracy and efficiency of business relationship chain scheduling.

[0005] In the first aspect, the present invention provides a method for automatically scheduling business relationship chains, and the method for automatically scheduling business relationship chains includes:

[0006] Construct a business rule library based on supplier information, customer information, and business information, and perform standardization processing on customer demand data to obtain standardized business demand information;

[0007] Perform multi-dimensional matching operations on the standardized business demand information based on the business rule library to obtain an initial supplier set, and perform dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence;

[0008] Construct a chain node graph based on the optimal supplier sequence, calculate and verify the business constraint relationships between nodes, and generate a business relationship chain;

[0009] Convert the business relationship chain into a standard order format, and complete order data synchronization and output an order synchronization result through an encryption signature and a concurrent distribution mechanism.

[0010] In the second aspect, the present invention provides a device for automatically scheduling business relationship chains, and the device for automatically scheduling business relationship chains includes:

[0011] A standardization processing module, configured to construct a business rule library based on supplier information, customer information, and business information, and perform standardization processing on customer demand data to obtain standardized business demand information;

[0012] A matching operation module, configured to perform multi-dimensional matching operations on the standardized business requirement information based on the business rule library to obtain an initial supplier set, and perform dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence;

[0013] A verification module, configured to construct a chain node graph based on the optimal supplier sequence, calculate and verify the business constraint relationships between nodes, and generate a business relationship chain;

[0014] An output module, configured to convert the business relationship chain into a standard order format, and complete order data synchronization and output an order synchronization result through an encryption signature and a concurrent distribution mechanism.

[0015] A third aspect of the present invention provides a computer-readable storage medium, in which instructions are stored, and when they run on a computer, the computer is made to execute the method for automatically arranging business relationship chains described above.

[0016] In the technical solution provided by the present invention, by establishing a multi-dimensional rule database and a standardized data processing mechanism, the management of business rules is made more systematic and standardized; by adopting a multi-dimensional matching operation and a dynamic weight adjustment strategy, the intelligent and accurate screening of suppliers is realized; by introducing the chain node graph technology, the visual management and automatic verification of business constraint relationships are carried out; by adopting an encryption signature and a concurrent distribution mechanism, the security and efficiency of order data synchronization are ensured; the system replaces manual operations through automated processing, reduces operation risks, improves the accuracy and efficiency of business relationship chain arrangement, and at the same time supports the dynamic adjustment and real-time response of business rules, enhancing the adaptability and scalability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a schematic diagram of the steps of the method for automatically arranging business relationship chains in the embodiments of the present invention;

[0019] Figure 2 It is a schematic diagram of the structure of the device for automatically arranging business relationship chains in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] An embodiment of the present invention provides a method, apparatus, and storage medium for automatically arranging business relationship chains. The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiment of the present invention will be described below. Please refer to Figure 1 , an embodiment of the method for automatically arranging business relationship chains in the embodiment of the present invention includes:

[0022] Step S1: Construct a business rule library based on supplier information, customer information, and business information, and perform standardization processing on customer demand data to obtain standardized business demand information;

[0023] It can be understood that the execution subject of the present invention can be a device for automatically arranging business relationship chains, or a terminal or a server. Specifically, it is not limited here. The embodiment of the present invention takes the server as the execution subject for illustration.

[0024] Specifically, classify and store data related to suppliers, customers, and business operations to ensure the rationality and scalability of the data structure. Split the supplier information into basic supplier information, qualification information, product type information, production capacity information, and rating information, and classify and store these data according to different attributes to form a supplier rule table. For example, the basic supplier information includes its name, address, contact information, registration number, etc., while the qualification information includes business licenses, industry certifications, compliance qualifications, etc. At the same time, the product type information is used to record the categories of products or services that the supplier can provide, and the production capacity information involves indicators such as production capacity and production cycle. The rating information is calculated comprehensively based on historical transaction data, customer feedback, and market evaluations. At the same time, classify and store customer information. The basic customer information includes company name, contact person, industry type, etc., while the credit rating information is calculated comprehensively based on historical transaction records, payment habits, overdue situations, etc. The payment terms information covers acceptable payment methods, payment periods, installment rules, etc., and the return rate information is used to measure the return ratio of customers in historical transactions to more accurately assess their credit risks. At the same time, the cooperation years information can reflect the historical cooperation between the customer and the existing suppliers in the system. These information are stored uniformly and a customer rule table is constructed. Classify and store the delivery time information, quality standard information, price range information, minimum order quantity information, and inventory warning information in business information to form a business rule table. Among them, the delivery time information covers the delivery cycle of orders, the average delivery time of suppliers, and the delivery time requirements of customers. The quality standard information involves industry quality certifications, product inspection reports, quality complaint records, etc. The price range information is used to define the market price fluctuation range to select the optimal price strategy during the supplier matching process. The minimum order quantity information stipulates the minimum purchase quantity for each transaction to ensure the economy and feasibility of the supply chain. At the same time, the inventory warning information is used to monitor the current inventory level of products and trigger corresponding replenishment strategies to ensure the stable operation of the business chain. Normalize the data in the supplier rule table, customer rule table, and business rule table to obtain standardized data. Based on the standardized data, establish cross-table field index relationships. By analyzing the field correlations between suppliers, customers, and business information, calculate the correlation strength between fields, and form an association index table accordingly. This process is based on various data analysis methods. For example, the Pearson correlation coefficient is used to measure the correlation between numerical fields, or the Jaccard similarity is used to analyze the similarity between categorical fields. For example, the matching degree between the product type of a supplier and the demand category of a customer is determined by calculating their overlap degree, and the relationship between the credit rating and payment terms is also analyzed based on historical data. Through these calculations, quantify the influence degree between different fields and optimize the accuracy of business rule matching based on the association index table.Update the trigger according to the associated index table setting rules to ensure the consistency and real-time nature of the rule base data. The role of the trigger is that when there are changes in suppliers, customers, or business rules, it automatically updates the relevant matching rules to avoid incorrect decisions caused by untimely data updates. For example, when the qualification information of a certain supplier changes, the trigger will automatically check whether the supplier still meets the requirements of the current customer. If not, it will be automatically removed from the matching results, and the supplier score will be recalculated to ensure that the business rule base always maintains the optimal configuration. At the same time, the rule update trigger is used to monitor risk signals in the supply chain in real time. For example, when the return rate of a certain customer suddenly increases, it automatically reduces the credit score of that customer and adjusts the corresponding trading conditions to reduce business risks. Collect and standardize customer demand data to obtain standardized business demand information to ensure that the requirements input by customers are consistent with the data format in the business rule base, thereby achieving efficient matching.

[0025] Deploy an event listener on the front-end interface through the Web collection unit. This listener monitors the demand data input by customers in the interface in real time, captures this data when the customers input, and obtains the interface input data. At the same time, select an API interface unit to use the RESTful architecture combined with the JSON Web Token (JWT) authentication mechanism to verify the demand data pushed by the external system. The design of the RESTful architecture enables the system to perform data interaction in a lightweight manner, while the JWT authentication mechanism ensures the credibility of the data source and avoids unauthorized systems from submitting false or malicious data. When the external system pushes data, the API interface unit parses the JWT token to verify the legality of the request, confirm whether the data comes from an authorized system, and then perform integrity verification on the received data, including field format verification, mandatory field check, and data consistency comparison, to ensure the accuracy of the external demand data. Identify duplicate data between the interface input data and the external demand data to eliminate redundant information and improve the efficiency of data storage and processing. Build an intelligent deduplication mechanism. By comparing the core fields of the data, such as customer ID, product type, delivery time, quantity, etc., judge whether the data is duplicate. In this process, use the hash algorithm to generate a unique fingerprint for the key fields, and quickly query through the database index whether there is data with the same fingerprint stored in the system, so as to accurately screen out duplicate data and ensure the uniqueness of the demand data. To improve the intelligence of the deduplication algorithm, introduce fuzzy matching technology, such as calculating the similarity of two demand data based on the edit distance algorithm to judge whether they belong to the same demand variant, and perform deduplication according to the similarity threshold. Perform intelligent filling on the deduplicated demand data to complete the missing fields and improve the integrity of the data. Infer the missing values based on historical transaction data and combine with machine learning models to predict the missing values by analyzing the demand patterns of similar customers to obtain the customer demand data. Perform field mapping conversion on the customer demand data according to the internal data format specification, establish a field mapping relationship, and convert the external data fields into the system internal field format. At the same time, during the field mapping process, perform data unit conversion to obtain demand data with a unified format. Based on the demand data with a unified format, set the processing priority according to the product type information and delivery requirement information to ensure that high-priority demands can be responded to faster and obtain standardized business demand information.

[0026] Step S2: Perform multi-dimensional matching operations on the standardized business demand information based on the business rule library to obtain an initial supplier set, and perform dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence;

[0027] Specifically, based on the product type information, suppliers with corresponding supply capabilities are screened from the business rule library. The product categories registered in the supplier rule table are queried and compared with the product types in the customer requirements to ensure that the products or services provided by the suppliers exactly match the customer requirements. At the same time, to improve the screening efficiency, database indexing is used to accelerate the query, or a method based on text similarity calculation is used to expand the search scope, allowing a certain degree of fuzzy matching to prevent the omission of suitable suppliers due to product naming differences, and a list of product-matching suppliers is obtained. Based on the customer rule table in the business rule library, black and white list filtering is performed on the supplier list to ensure that only suppliers meeting the customer's cooperation requirements are retained. The cooperation qualifications of the suppliers are reviewed to check whether the suppliers are on the customer's blacklist. If there are historical disputes, quality problems or credit risks, they are immediately excluded. Secondly, white list suppliers are preferentially retained to ensure that suppliers with long-term stable cooperation are preferentially matched. At the same time, for suppliers that are not clearly classified, further screening is carried out in combination with historical transaction data and market reputation evaluation to enhance the flexibility and accuracy of the matching, and a list of suppliers with cooperation qualifications is obtained. Based on the delivery time requirements in the customer requirements, the delivery capabilities of the suppliers with cooperation qualifications are verified to ensure that the suppliers can deliver the required products on time. The historical delivery records of the suppliers are queried, and their delivery capabilities are calculated in combination with the current order backlog situation. For example, by analyzing the average delivery cycle, historical order fulfillment rate and recent production schedules of the suppliers, it is evaluated whether they can meet the delivery time limit of the current order. At the same time, logistics factors are considered, such as the geographical location, transportation mode and inventory level of the suppliers, in order to more accurately predict the delivery time. If the expected delivery cycle of a certain supplier exceeds the time required by the customer, it will be excluded, and a list of suppliers with qualified deliveries is obtained. According to the order size requirements, the production capacity matching degree of the suppliers is calculated to ensure that the suppliers have sufficient production capacity to meet the customer requirements. The maximum production capacity, current order volume and production schedule of the suppliers are checked, and the matching degree between their available production capacity and the order size is calculated. The calculation method used is based on the production capacity utilization rate of the suppliers. For example, by calculating their current production capacity load rate, if the load rate is too high, it will lead to delivery delays and then reduce the matching priority. For suppliers with sufficient production capacity and flexible deployment capabilities, they are preferentially included in the next stage of screening to form a list of suppliers with matching production capacity. The suppliers in the list of suppliers with matching production capacity are scored for credibility according to their historical cooperation records to evaluate the reliability of the suppliers. The calculation of the credibility score synthesizes multiple indicators, including customer feedback score, historical order fulfillment rate, return rate, complaint record, etc., and the comprehensive credibility score is calculated through weighted calculation. In order to optimize the stability of the supply chain, the production capacity balance calculation is carried out in combination with the current order reception volume of the suppliers to ensure that those suppliers with both high credibility and no overloaded operation are preferentially selected, and to avoid affecting the delivery quality due to production capacity overload, and an initial supplier set is obtained.Calculate the comprehensive score of the initial supplier set, and dynamically adjust the weights of the supplier scores to obtain the optimal supplier sequence. Consider multiple dimensions such as price, delivery time, credit score, production capacity matching degree, etc., and dynamically adjust the weights according to the specific needs of customers. For example, if the customer has high requirements for delivery time, the weight of delivery capacity will be increased, and if the customer pays more attention to cost control, the price weight will be given priority. At the same time, combine the market supply and demand situation and historical transaction data to dynamically optimize the scoring weights to adapt to the changes in the business environment. Finally, generate the optimal supplier sequence through multi-dimensional comprehensive calculation.

[0028] Build a supplier state space model based on the initial supplier set. By vectorizing the production capacity status, quality status, and price status of suppliers, convert the discrete business characteristics into high-dimensional numerical vectors to form supplier state vectors. Construct a reasonable state representation. Among them, the production capacity status includes information such as the maximum production capacity, current production capacity load rate, and expandable production capacity of the supplier. The quality status covers dimensions such as historical order quality scores, customer feedback, and return rates. The price status considers factors such as market price fluctuations, the deviation between the supplier's quotation and the industry average price, and long-term price trends. Through feature standardization and normalization methods, all state features are calculated on the same scale to form supplier state vectors. Model the action space for the supplier state vector and business objectives. That is, in the process of supplier selection and optimization, take the three key factors of price adjustment, production capacity allocation, and quality control as action dimensions to establish an action space model. The role of price adjustment is to balance costs and market competitiveness. By adaptively adjusting the supplier's quotation, optimize the overall supply chain cost. Production capacity allocation involves optimizing the allocation of the supplier's production capacity to ensure that high-quality suppliers match the most suitable orders within a reasonable production capacity range, thus avoiding supply chain bottlenecks. Quality control is used to monitor and improve the product or service quality of suppliers. For example, through means such as order acceptance mechanisms and historical data analysis, ensure that the selected suppliers meet quality standards. Model these action dimensions as part of the optimization strategy to more effectively adjust the supplier selection plan to meet the needs of different business objectives. After establishing the action space model, perform feature extraction and dimensionality reduction processing on the supplier state vector to reduce the computational complexity and improve the optimization efficiency. Use principal component analysis, autoencoders, or other dimensionality reduction algorithms to compress the state data and extract the most representative supplier state feature matrix. Input the dimensionality-reduced state feature matrix into the value network for iterative calculation, and evaluate the different action values in combination with historical business data to obtain the action value matrix. The value network models the supplier decision-making process through deep reinforcement learning (such as the deep Q-network DQN or Actor-Critic method), and calculates the long-term value of different actions using historical order data and market feedback data. For example, whether adjusting the supplier's price can increase the overall profit, or whether optimizing the production capacity allocation can reduce the supply chain risk, so as to construct a more accurate supplier optimization strategy. Sample and simulate the supplier chain based on the action value matrix to construct different supplier combination plans and analyze their business feasibility. Use the Monte Carlo method or Markov decision process to sample the supplier chain, simulate the performance of different supplier combinations under various market conditions, and calculate the policy probability distribution based on the sampling results. The policy probability distribution is used to measure the optimal probability of each supplier combination and determine the direction of supplier selection.According to the policy probability distribution, an exploratory search is conducted on suppliers to find undervalued but potentially competitive suppliers, and the search results are fed back to the value network for parameter update, thus optimizing the overall action strategy. During the search process, Bayesian optimization or evolutionary algorithms are combined to make the search process balance local optimization and global exploration and avoid falling into local optima, forming a more stable and effective supplier optimization strategy. After the search results are fed back to the value network, the value network recalculates the action values based on the new data and updates the neural network parameters to make the optimization strategy more in line with the actual business needs. After optimizing the action strategy, the optimized action strategy is used to predict the transition of the supplier state, that is, based on the existing supplier state, the future state changes are predicted, and a state transition sequence is generated accordingly. The state transition sequence describes the possible state evolution of suppliers in the future business environment. For example, the production capacity of a certain supplier is adjusted with the change of market demand, the price changes due to the fluctuation of raw material costs, and the quality status is improved due to the improvement of production technology. During the state transition prediction process, time series prediction models (such as LSTM or ARIMA) are combined to improve the prediction accuracy and ensure the foresight of supplier optimization decisions. After the state transition sequence is constructed, the cumulative reward value is calculated based on the preset reward function to evaluate the long-term benefits of different supplier combinations. The setting of the reward function is based on business objectives, such as minimizing the supply chain cost, maximizing the order fulfillment rate, optimizing customer satisfaction, etc., and the cumulative reward value is used to measure the overall advantages and disadvantages of different supplier selection options and determine the optimal state path accordingly. The optimal state path represents the optimal supplier decision sequence under long-term business planning. The initial supplier set is re-sorted and screened according to the optimal state path, and a candidate supplier sequence is generated accordingly as the input data for comprehensive score calculation.

[0029] Obtain the candidate supplier sequence calculated from the initial supplier set, and based on this sequence, conduct multi-dimensional data collection on core indicators such as the price competitiveness, quality stability, and delivery timeliness of suppliers to construct a supplier evaluation data set. During the data collection process, the measurement of price competitiveness needs to be based on the historical quotations of suppliers, the market average price, cost fluctuations, and the price adjustment mechanism in long-term cooperation agreements. Quality stability involves the historical order qualification rate, product failure rate, customer complaint rate, and quality certification information of suppliers. Delivery timeliness mainly depends on the performance rate of suppliers, delivery delay situations, logistics efficiency, and supply chain reliability. By integrating these data, construct an evaluation data set reflecting the business capabilities of suppliers. Allocate weights to the various evaluation indicators of suppliers according to the evaluation rule table in the business rule library, and calculate the initial weight coefficients. The evaluation rule table provides the default weights for different evaluation indicators in supply chain management, and these weights are obtained based on industry standards, enterprise strategies, and historical data analysis. By setting different initial weight coefficients, ensure that the scoring system conforms to the business logic of the enterprise. Calculate the importance of historical performance data involved in the supplier evaluation data set to construct an index weight matrix, and on this basis, dynamically adjust the initial weight coefficients to obtain optimized weight coefficients. Adopt a variety of mathematical methods to calculate the importance of each indicator. For example, perform dimensionality reduction on the data based on principal component analysis to extract the most influential key indicators, or use the entropy weight method to calculate the information entropy of each indicator, and adjust the weights through the method of information gain, so that indicators with larger fluctuations obtain higher weights, while the weights of relatively stable indicators are appropriately reduced. Combine historical transaction data, calculate the influence of different indicators on the final supplier performance through regression analysis or machine learning models, form a more accurate weight allocation plan, enable the scoring system to dynamically adapt to market changes, and ensure the rationality of supplier selection. Calculate the weighted average of the supplier evaluation data set according to the optimized weight coefficients to calculate the comprehensive score of each supplier, and rank the suppliers in descending order based on this comprehensive score to obtain the optimal supplier sequence.

[0030] Step S3: Construct a chain node graph based on the optimal supplier sequence, calculate and verify the business constraint relationships between nodes, and generate a business relationship chain;

[0031] Specifically, a unique chain identification code is generated based on the optimal supplier sequence. Through a combined operation of a timestamp and a random string, the uniqueness and traceability of the chain number are ensured. A high-precision timestamp (such as a millisecond-level or nanosecond-level timestamp) is used to record the time when the chain is generated, and a unique identifier is generated through a random string or a hash function, thereby avoiding number conflicts and ensuring that each specific chain instance can be accurately located during the management and tracking of the business chain. After generating the chain number, the supplier information, customer information, and business rule information in the optimal supplier sequence are processed into nodes, and an independent node identifier and node attributes are assigned to each node to form the initial node set of the chain. Each supplier, customer, and involved business rule is regarded as an independent node and is given corresponding attributes. For example, the supplier node includes the supplier number, product category, production capacity information, price, etc., while the customer node includes the customer ID, credit rating, payment method, required products, etc., and the business rule node records business constraints such as payment conditions, delivery time, and quality requirements. By structuring this information into independent nodes, it is ensured that various business elements can be stored and processed in a standardized manner during the subsequent establishment of the relationship chain. The nodes in the initial node set of the chain are connected with directed edges according to the business flow to construct a chain constraint graph. Each supplier node is connected to the corresponding customer node, and the direction of the connection relationship depends on the business logic. For example, the order flow is initiated by the customer, so the data flow is from the customer node to the supplier node, while the payment process is the opposite and requires payment from the customer to the supplier, so a reverse connection from the supplier to the customer is established. Business rule constraints, including payment conditions, delivery time, inventory availability, etc., are marked on each connection edge, thereby ensuring that the business dependence relationship and transaction rules between each node are reflected during the subsequent business execution process. After completing the construction of the chain constraint graph, conflict detection is performed on the business rules to ensure that all business relationships between nodes comply with the preset rules. A graph traversal algorithm (such as depth-first search DFS or breadth-first search BFS) is used to verify the integrity of the graph and check for rule conflicts or circular dependencies. For example, if the delivery cycle of a certain supplier cannot meet the customer's requirements, a rule conflict will occur between the supplier node and the customer node, and if multiple suppliers involved in a certain business chain form a dependency loop, it will cause business blockage, and these problems must be detected and corrected at this stage. During the traversal process, each business rule is verified as a constraint condition, and all detected constraint conflicts are recorded to form verification result data. Based on the verification result data, the chain constraint graph is optimized and adjusted, abnormal nodes are corrected, and the constraint relationship is recalculated to generate the final target chain graph.In this process, based on the automatic rule adjustment algorithm, for example, in the case of delivery cycle conflicts, the supplier priorities are dynamically adjusted to select suppliers that better meet the delivery requirements, or additional suppliers are introduced when necessary to share the order demand. For the problem of circular dependencies, the loop paths are split based on the topological sorting method to ensure that the dependency relationships between all nodes conform to the business logic. After the optimization of the target chain graph is completed, an associated index is established with the business rule library, so as to ensure that the business chain can dynamically adapt to the updates of the rule library and quickly adjust the chain structure when the future business environment changes. This process is achieved through database indexing or a relationship indexing mechanism based on a graph database, enabling each chain instance to be directly associated with the data in the rule library and supporting fast query and rule matching. The generated business relationship chain not only has complete business logic and constraint rules, but also ensures rapid optimization and adjustment in case of changes in the business environment through the indexing mechanism, thereby enhancing the flexibility, stability, and scalability of the business relationship chain.

[0032] Step S4: Convert the business relationship chain into a standard order format, and complete the order data synchronization and output the order synchronization result through an encryption signature and a concurrent distribution mechanism.

[0033] Specifically, extract customer information, supplier information, product information, and delivery requirement information from the business relationship chain, and perform format conversion according to the preset order field mapping rules to ensure compatibility and data standardization between different systems. Customer information includes customer number, contact person, credit rating, payment terms, etc., while supplier information covers supplier number, product category, production capacity, quotation information, etc. Product information involves SKU number, product specification, unit price, order quantity, etc., and delivery requirement information includes delivery time, delivery location, logistics method, acceptance criteria, etc. Through the conversion of the field mapping rules, ensure that all order data conforms to the standard format and can be seamlessly docked with the downstream order management system to form initial order data. Encrypt the initial order data to ensure the security and integrity of the order data during transmission. Use the SHA-256 algorithm to perform a digital signature on the order data to generate a unique verification value for the order, and ensure that the order content has not been tampered with, which is also used for subsequent data integrity verification. Encrypt sensitive fields in the order data (such as customer contact information, supplier payment account, order amount, etc.) using the AES-256 algorithm to prevent the data from being stolen or tampered with by unauthorized third parties during transmission, generating encrypted order data. According to the system type of the order recipient, select the appropriate communication protocol and data format from the protocol library, and perform protocol encapsulation on the encrypted order data to ensure that different business systems can correctly parse and process the order data. Select the appropriate protocol according to the technical requirements of the order recipient and perform data format conversion. Package the data using standard formats such as JSON, XML, CSV, etc., and ensure that the field structure conforms to the specifications of the recipient system to ensure the compatibility and readability of the order data, forming standard order format data. Input the standard order format data into a thread pool for concurrent task allocation, and establish independent communication channels for the order data of different recipients to improve the transmission efficiency of the order data. Based on the thread pool management mechanism, batch process different orders, and perform dynamic task scheduling according to factors such as order priority, target system response time, network load, etc., so as to maximize the use of computing resources and ensure that the order is transmitted to the recipient at the fastest speed. At the same time, the transmission process of each order is carried out in an independent communication channel to avoid interference between different orders and ensure high availability of the data during transmission, forming a concurrent processing queue. During the transmission of the order data, monitor the sending status of the orders in the concurrent processing queue in real time to ensure that all orders can be successfully transmitted and confirmed by the recipient. The system records the sending time, received confirmation information, and error logs for each order, and judges the transmission status of the order through log analysis. If the order data is successfully transmitted and the recipient returns a confirmation message, record the successful order synchronization and update the order status; if the recipient fails to correctly receive the order, the system will automatically mark the order as "synchronization failed" and trigger the corresponding retry mechanism.The retry strategy conducts intelligent scheduling based on the exponential backoff algorithm. That is, after the first failure, a retry will be performed within a short period of time. If consecutive failures occur, the retry interval will be gradually extended to reduce network load and increase the retry success rate. If it still fails after multiple retries, the manual intervention process will be triggered to notify the relevant operators to manually check the cause of the failure and provide error logs and data records, so that the operation and maintenance team can quickly locate the problem and adopt appropriate solutions, such as manually supplementing orders, adjusting data formats, repairing network connections, etc., ultimately ensuring that all orders can be correctly transmitted and obtaining the order synchronization result.

[0034] In the embodiments of the present invention, by establishing a multi-dimensional rule database and a standardized data processing mechanism, the management of business rules becomes more systematic and standardized; by adopting multi-dimensional matching operations and dynamic weight adjustment strategies, the intelligent and precise screening of suppliers is realized; by introducing the chain node graph technology, the visualization management and automatic verification of business constraint relationships are carried out; by adopting the encryption signature and concurrent distribution mechanism, the security and efficiency of order data synchronization are ensured; the system replaces manual operations with automated processing, reduces operation risks, improves the accuracy and efficiency of business relationship chain orchestration, and at the same time supports the dynamic adjustment and real-time response of business rules, enhancing the adaptability and scalability of the system.

[0035] In a specific embodiment, the process of executing step S1 may specifically include the following steps:

[0036] Classify and store the supplier basic information, qualification information, product type information, production capacity information, and rating information in the supplier information to obtain a supplier rule table;

[0037] Classify and store the customer basic information, credit rating information, payment terms information, return rate information, and cooperation years information in the customer information to obtain a customer rule table;

[0038] Classify and store the delivery time information, quality standard information, price range information, minimum order quantity information, and inventory warning information in the business information to obtain a business rule table;

[0039] Normalize the data in the supplier rule table, customer rule table, and business rule table to obtain standardized data;

[0040] Based on the standardized data, establish a cross-table field index relationship, calculate the association strength between fields to obtain an association index table, and set a rule update trigger according to the association index table to verify and update the consistency of the rule data to obtain a business rule library;

[0041] Collect customer demand data and perform standardized processing to obtain standardized business demand information.

[0042] Specifically, the supplier information is refined and classified, and the supplier's basic information, qualification information, product type information, production capacity information, and rating information are stored in a unified data structure to form a supplier rule table. Among them, the supplier's basic information includes the supplier's unique number S1, enterprise name S2, registered address S3, contact information S4, and registered capital S5, which are used to identify the basic attributes of the supplier. The qualification information includes enterprise qualification certification S6, tax registration number S7, industry certification level S8, and quality management system certification S9, which are used to evaluate the compliance of the supplier. The product type information S 10 records the product categories provided by the supplier, stores the SKU (Stock Keeping Unit) numbers in a standardized manner, and at the same time, the production capacity information S 11 reflects the maximum production capacity, production capacity utilization rate, and production cycle of the supplier, while the rating information S 12 is comprehensively calculated from historical transaction data, customer feedback, and market evaluations to measure the credibility and long-term stability of the supplier. When constructing the customer rule table, it is stored according to customer basic information, credit rating information, payment terms information, return rate information, and cooperation years information. The customer basic information includes customer number C1, enterprise name C2, industry category C3, main contact person C4, and contact information C5, which are used to identify the customer's identity. The credit rating information C6 is calculated based on historical transaction data, for example, based on the Bayesian scoring model:

[0043]

[0044] where, T i is the credit score of the i-th transaction, W i is the weight of this transaction, and n is the number of the customer's historical transactions. The payment terms information C7 records the customer's payment method (such as prepayment, installment payment, credit period), and the calculation method of the return rate information C8 is:

[0045]

[0046] where, R represents the number of orders returned by the customer, and O represents the total number of orders of the customer. The cooperation years information C9 records the length of time the customer has cooperated with the enterprise. At the same time, the business information is classified and stored according to delivery time information, quality standard information, price range information, minimum order quantity information, and inventory warning information to form a business rule table. The delivery time information B1 records the historical delivery times of different suppliers, and the quality standard information B2 sets the industry quality inspection indicators, such as the product defect rate B 21 is calculated as:

[0047]

[0048] Among them, D is the quantity of defective products, and Q is the total production quantity. The price range information B3 records the lowest price, market average price, and highest price of the product. The minimum order quantity information B4 sets the minimum purchase quantity acceptable to the supplier, while the inventory warning information B5 sets the replenishment threshold based on the real-time inventory level. Normalize the supplier rule table, customer rule table, and business rule table to eliminate differences in data formats from different sources and convert the values to the same scale. For example, normalize the supplier rating S 12 using min-max normalization:

[0049]

[0050] where and are the maximum and minimum values of the supplier rating respectively. After completing the normalization process, establish cross-table field index relationships based on the standardized data and calculate the association strength between fields to construct an association index table. Suppose a supplier S i provides a certain product P j to a certain customer C k , calculate the association strength R between the supplier and the customer:

[0051] R ik = α1·F1 + α2·F2 + α3·F3;

[0052] where F1 is the historical transaction times between the supplier and the customer, F2 is the on-time delivery rate of the supplier, F3 is the credit score of the customer, and α1, α2, α3 are weight coefficients, which are optimized through empirical data or machine learning methods (such as gradient boosting trees). Based on these calculations, generate an association index table to establish data associations between suppliers, customers, and business information. After the association index table is constructed, set a rule update trigger to ensure that when the business rules change, the system automatically adjusts the relevant data. For example, when the qualification information of a certain supplier is updated, trigger an automatic verification process to check whether it still meets the business requirements of the current customer and update the data in the supplier rule table to ensure data consistency. After constructing a complete business rule library, collect customer demand data and perform standardization processing to ensure that its format is consistent with the data in the business rule library. The customer demand data D includes product type D1, purchase quantity D2, delivery time D3, payment method D4, and supplier preference D5. For matching, perform mapping conversion on the customer demand data. For example, convert the product name entered by the customer into a standard SKU number, and complete the data based on the customer's historical purchase records. For example, if the customer does not fill in the delivery time, use its historical average delivery cycle T avg :

[0053]

[0054] Among them, T i is the historical delivery cycle data of the customer, and n is the number of historical transactions. Through the standardized customer demand data, it is efficiently matched with the business rule base to achieve the automated orchestration of the business relationship chain and ensure the consistency and optimization of the entire supply chain data.

[0055] In a specific embodiment, the process of performing the steps of collecting customer demand data and performing standardized processing to obtain standardized business demand information may specifically include the following steps:

[0056] The Web collection unit uses an event listener to capture the demand data input by the customer on the interface in real time to obtain the interface input data;

[0057] The API interface unit uses the RESTful architecture combined with the JSON Web Token authentication mechanism to verify the demand data pushed by the external system to obtain the external demand data;

[0058] Identify duplicate data in the interface input data and the external demand data to obtain the deduplicated demand data, and intelligently fill in the missing fields in the deduplicated demand data to obtain the customer demand data;

[0059] Perform field mapping conversion on the customer demand data according to the internal data format specification to obtain the demand data with a unified format, and set the processing priority according to the product type information and delivery requirement information in the demand data with a unified format to obtain the standardized business demand information.

[0060] Specifically, deploy a JavaScript-based event listener on the front-end page. This listener listens in real time for interaction events such as user input boxes, drop-down options, and file uploads, so as to immediately capture the input content when the user inputs or modifies data. For example, when inputting product requirements, when the keyup event is triggered, the listener records the current input product number D1, purchase quantity D2, delivery time D3, and payment method D4, and submits the data to the backend through a WebSocket or Ajax request to ensure the real-time and integrity of the data. At the same time, the API interface unit uses the RESTful architecture and combines the JSON Web Token (JWT) authentication mechanism to verify the demand data pushed by the external system to ensure the security and reliability of the data source. When each external system submits data, it must carry a JWT token, which contains the system identity information T1 and the authorization scope T2, and performs signature verification through HMAC-SHA256 encryption. After receiving the request, the server parses the JWT token and calculates its hash value H:

[0061] H = HMAC SHA256 (T1||T2, K);

[0062] Where K is the private key on the server side, || represents string concatenation. If the calculated H matches the signature in the request, it proves that the request is legal; otherwise, the processing is refused. Meanwhile, the server parses the product type D5, order size D6, delivery time D7, and customer number D8 in the external demand data, and conducts data format verification to ensure that the data conforms to the system specifications, obtaining the verified external demand data. Duplicate data identification is performed on the interface input data and the external demand data to remove redundant information and improve data quality. The deduplication process uses a similarity matching algorithm based on hash values to calculate the unique hash value H of each demand data i :

[0063] H i = SHA256(D1||D2||D3||D4);

[0064] All hash values are stored in a hash table, and the data list is traversed. If data with the same hash value is found, it is determined as duplicate data, and only the latest one is retained, forming the deduplicated demand data. Intelligent filling is performed on the deduplicated demand data to complete incomplete data. The intelligent filling method makes inferences based on historical data. For example, for the missing delivery time D3, calculate the historical average delivery time T of the customer avg :

[0065]

[0066] Where T i represents the past delivery time records of this customer, and n is the number of historical orders. If the historical data is insufficient, the industry standard delivery time T std is used as the default value. For the missing payment method, it is filled based on the customer's credit rating C6. For example:

[0067]

[0068] Automatically select an appropriate payment method according to the customer's credit situation, thereby improving the rationality of order matching. After completing the data filling, field mapping conversion is performed on the customer demand data according to the internal data format specifications to ensure that all data structures are unified and meet the requirements of the business system. Similarly, unit conversion is performed to ensure that all data units are unified to avoid calculation errors. Set the processing priority according to the product type information and delivery requirement information in the demand data with unified format, thereby optimizing the response speed of order matching. The processing priority calculation formula is as follows:

[0069] P = α1×W1 + α2×W2;

[0070] Among them, W1 is the urgency level of the order, such as whether it is an urgent order, W2 is the customer credit score, and α1 and α2 are weight coefficients determined by business requirements. All processed data is converted into standardized business requirement information and stored in the database for use in the automatic orchestration of subsequent business relationship chains.

[0071] In a specific embodiment, the process of executing step S2 may specifically include the following steps:

[0072] Screen out suppliers with supply capabilities from the business rule library according to the product type information in the standardized business requirement information to obtain a list of product-matching suppliers;

[0073] Filter the list of product-matching suppliers based on the customer rule table in the business rule library to obtain a list of suppliers with cooperation qualifications;

[0074] Verify the delivery capabilities of the list of suppliers with cooperation qualifications according to the delivery time requirements in the standardized business requirement information to obtain a list of suppliers that meet the delivery standards;

[0075] Calculate the production capacity matching degree of the list of suppliers that meet the delivery standards according to the order scale requirements in the standardized business requirement information to obtain a list of suppliers with matching production capacity;

[0076] Score the suppliers in the list of suppliers with matching production capacity according to their historical cooperation records to obtain a list of credit scores, and perform supplier production capacity balance calculation based on the list of credit scores combined with the current order acceptance volume of the suppliers to obtain an initial supplier set;

[0077] Perform comprehensive score calculation and dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence.

[0078] Specifically, extract the product type information from the standardized business requirement information, and then query the suppliers with the supply capacity of this product in the business rule library to form a list of product-matching suppliers. In this process, for each supplier S i whose supply capacity C i is determined by its product type P i and historical supply records H i , that is:

[0079] C i = f(P i , H i );

[0080] Among them, P i represents the product categories that the supplier can provide, and H iRepresents the past supply success rate of the supplier. For example, if a supplier has successfully fulfilled 9 out of the past 10 orders, its supply success rate is 90%. When screening, the system preferentially selects suppliers with a supply success rate higher than a certain threshold H min of suppliers:

[0081] S i ∈{S|C i ≥H min};

[0082] Based on the customer rule table, black and white list filtering is performed to ensure that the selected suppliers meet the customer's cooperation requirements. The customer rule table R c records the list of cooperative suppliers for each customer C k , including the blacklist B k and the whitelist W k , and the screening rule is:

[0083]

[0084] The supplier must be on the customer's whitelist or at least not on the blacklist. Through this step, a list of suppliers with cooperation qualifications is formed. According to the customer's delivery time requirements, the suppliers with cooperation qualifications are screened to ensure that they can deliver products on time. The delivery ability D i of the supplier is determined by comparing its average delivery cycle T i with the customer's delivery deadline T c :

[0085] D i =T c -T i ;

[0086] If D i ≥0, it means that the supplier can deliver on time, otherwise the supplier will be excluded. All suppliers who meet the delivery time requirements form a list of suppliers that meet the delivery standards. After ensuring that the supplier has the ability to deliver on time, consider whether the supplier can meet the order size requirements and calculate its production capacity matching degree. The available production capacity P i of the supplier needs to meet the customer's order demand Q c :

[0087] P i ≥Q c ;

[0088] If the current production capacity P i of the supplier is less than the order demand, the supplier will be excluded. Suppliers who meet the order size requirements form a list of suppliers with matching production capacity. After screening out suppliers with sufficient production capacity, the suppliers are rated for their credibility to measure their performance in past cooperation. The credibility score Ri From the historical cooperation score H i , the customer complaint rate C i and the order fulfillment rate O i are comprehensively calculated as follows:

[0089] R i = w1H i - w2C i + w3O i ;

[0090] Among them, w1, w2, w3 are the weight coefficients of different scoring dimensions. After calculating the credit score, combined with the current order volume L of the supplier i a production capacity balance calculation is carried out to ensure that the delivery will not be affected by the order overload of the supplier. The available production capacity of the supplier is calculated as follows:

[0091] P′ i = P i - L i ;

[0092] If P′ i is too low, the priority of this supplier will be reduced to obtain the initial supplier set. A comprehensive score calculation is carried out on the initial supplier set, and the dynamic weight of the supplier is adjusted to obtain the final optimal supplier sequence. The comprehensive score S i is calculated by weighting multiple factors:

[0093] S i = α1R i + α2P′ i + α3D i - α4C i ;

[0094] Among them, α1, α2, α3, α4 are the weight coefficients set according to business requirements. All suppliers are sorted according to the comprehensive score, and finally the optimal supplier sequence is formed.

[0095] Among them, before verifying the delivery capabilities of the list of cooperative qualified suppliers according to the delivery time requirements in the standardized business requirement information, it also includes: constructing a multi-frequency spatio-temporal feature model based on the historical delivery data of suppliers, including: extracting the real-time delivery data, daily delivery data, and weekly delivery data of suppliers from the business rule library, stratifying the data in the time dimension to obtain a multi-frequency time series dataset; performing time feature decomposition on the multi-frequency time series dataset, extracting the periodic change features of the suppliers' delivery capabilities to obtain time feature vectors; constructing a geographical location relationship graph of suppliers, converting information such as the logistics distance and warehouse layout between suppliers into an adjacency matrix to obtain a spatial feature map; constructing an adaptive spatial feature extraction module based on the spatial feature map, modeling the geographical dependence relationship between suppliers to obtain spatial feature vectors; collecting external event information, including factors such as weather conditions, traffic conditions, and holidays, establishing an event feature library, and performing correlation analysis on the event features and the historical performance data of suppliers to obtain an event impact model; fusing the time feature vectors, spatial feature vectors, and event impact model, constructing a multi-layer graph neural network, modeling the comprehensive delivery capabilities of suppliers to obtain a delivery capability prediction model; using the delivery capability prediction model to predict the future delivery performance of suppliers, generating delivery capability prediction results; classifying and grading suppliers according to the delivery capability prediction results, establishing a dynamic delivery capability evaluation system, and providing decision-making support for subsequent delivery capability verification. Specifically, performing time feature decomposition based on the multi-frequency time series dataset includes: using a time series decomposition algorithm to decompose the suppliers' delivery data into a trend term, a seasonal term, and a random term to obtain decomposition features; performing long-term change trend analysis on the trend term in the decomposition features, using a sliding window method to extract trend features to obtain trend feature vectors; performing periodic pattern recognition on the seasonal term in the decomposition features, using Fourier transform to extract periodic features to obtain periodic feature vectors; performing volatility analysis on the random term in the decomposition features, calculating volatility indicators to obtain volatility feature vectors; fusing the trend feature vectors, periodic feature vectors, and volatility feature vectors to obtain time feature vectors. Optionally, constructing an adaptive spatial feature extraction module based on the spatial feature map includes: constructing a multi-layer graph attention network to model the spatial dependence relationship between supplier nodes; designing an adaptive weight update mechanism to dynamically adjust the attention weights according to the business association degree between nodes; introducing a spatial convolutional layer to extract the geographical distribution features of suppliers; designing a residual connection structure to retain the original spatial feature information and prevent feature information loss; aggregating the multi-layer features to obtain the final spatial feature vector.

[0096] In a specific embodiment, the process of performing step to calculate the comprehensive score and dynamically adjust the weights of the initial supplier set to obtain the optimal supplier sequence may specifically include the following steps:

[0097] Collect multi-dimensional data on suppliers in the initial supplier set according to price competitiveness, quality stability, and delivery timeliness to obtain a supplier evaluation data set;

[0098] Allocate weights to the supplier evaluation data set according to the evaluation rule table in the business rule library to obtain the initial weight coefficients;

[0099] Calculate the importance of indicators for the historical performance data in the supplier evaluation data set to obtain an indicator weight matrix, and dynamically adjust the initial weight coefficients based on the indicator weight matrix to obtain optimized weight coefficients;

[0100] Perform a weighted average calculation on the supplier evaluation data set according to the optimized weight coefficients to obtain a comprehensive supplier score, and sort the suppliers in descending order based on the comprehensive supplier score to obtain an optimal supplier sequence.

[0101] Specifically, extract relevant data of suppliers from the business rule library and analyze historical transaction information. The price competitiveness of a supplier is compared by its quoted price P1 and the market average price P2, and its price competition index C1 is calculated:

[0102]

[0103] Among them, P1 represents the product quoted price of the supplier, and P2 represents the market average price of all suppliers for the same product. If C1 is positive, it means that the supplier's quoted price is lower than the market average price and the price competitiveness is strong; if it is negative, the competitiveness is weak. At the same time, the quality stability of the supplier is evaluated through historical order data, expressed by the qualification rate Q1:

[0104]

[0105] Among them, O1 represents the number of orders successfully delivered by the supplier and meeting the quality standards, and O2 represents the total number of orders of the supplier. The higher the qualification rate, the stronger the quality stability of the supplier. The delivery timeliness of the supplier is calculated by the average delay time T1:

[0106]

[0107] Among them, T2 and T3 respectively represent the actual delivery time and the estimated delivery time of the i-th order, and n is the total number of historical orders of this supplier. If T1 > 0, it indicates that the supplier's deliveries are generally delayed. If T1 ≤ 0, it shows that the supplier has strong delivery capabilities. Based on the above three key indicators, a supplier evaluation dataset is constructed, which includes the price competitiveness C1, quality stability Q1, and delivery timeliness T1 of each supplier. After obtaining the supplier evaluation dataset, weight distribution is carried out on each indicator according to the evaluation rule table in the business rule library to obtain the initial weight coefficients. Assuming that the weights of price competitiveness, quality stability, and delivery timeliness are W1, W2, and W3 respectively, the initial weight matrix is expressed as:

[0108] W = [W1, W2, W3];

[0109] Among them, the initial values of the weights are set according to business requirements and market conditions to ensure that the comprehensive score of the supplier reasonably reflects its business capabilities. Calculate the importance of the indicators in the supplier evaluation dataset to generate an indicator weight matrix, and dynamically adjust the initial weight coefficients based on this matrix. Use the entropy weight method to calculate the information entropy E1 of each indicator:

[0110]

[0111] Among them, p1 represents the normalized value of the i-th supplier on the first indicator, and m is the number of suppliers. The larger the entropy value, the lower the information increment of this indicator, and its importance should be reduced. Therefore, calculate the adjusted weight correction coefficient:

[0112]

[0113] The optimized weight coefficient is obtained by weighted average adjustment:

[0114] W5 = βW1 + (1 - β)W4;

[0115] Among them, β is used as a balance coefficient, generally taking values between [0, 1] to balance the original business rule weights and the results of data-driven weight calculations. After obtaining the optimized weight coefficient, perform a weighted average calculation on the supplier evaluation dataset to calculate the comprehensive score S1 of each supplier:

[0116] S1 = W5C1 + W6Q1 - W7T1;

[0117] Among them, C1 represents price competitiveness, Q1 represents quality stability, and T1 represents the number of delivery delay days. All suppliers are sorted in descending order of the comprehensive score to obtain the optimal supplier sequence.

[0118] Further, before constructing the chain node map based on the optimal supplier sequence, it also includes: converting the business collaboration relationships between suppliers in the optimal supplier sequence into a relationship matrix, quantitatively calculating the collaboration intensity to obtain an initial relationship intensity matrix; setting a business association benchmark threshold based on the initial relationship intensity matrix, and using an adaptive method to calculate the relationship determination threshold to obtain a 0-1 binary determination criterion; performing binary processing on the initial relationship intensity matrix according to the 0-1 binary determination criterion to mark valid business associations and obtain a binary relationship matrix; using a sliding window method to perform temporal accumulation on the association relationships in the binary relationship matrix, calculating the stable collaboration degree to obtain a collaboration accumulation matrix; identifying abnormal associations and missing associations in the supplier sequence based on the collaboration accumulation matrix to obtain a list of association optimization suggestions; fine-tuning the supplier sequence according to the list of association optimization suggestions, supplementing missing associations, and removing false associations to obtain an optimized optimal supplier sequence; verifying the association effectiveness of the optimized optimal supplier sequence, calculating the overall collaboration index to obtain a collaboration verification result; evaluating the structural stability of the supplier sequence based on the collaboration verification result, setting structural optimization parameters, and obtaining the final optimal supplier sequence.

[0119] In a specific embodiment, the process of executing step S3 may specifically include the following steps:

[0120] Generate a unique chain identification code based on the optimal supplier sequence, combine the timestamp with a random string for arithmetic operation to obtain a chain number;

[0121] Node-process the supplier information, customer information, and business rule information in the optimal supplier sequence, and assign a node identifier and node attributes to each node to obtain an initial chain node set;

[0122] Connect the nodes in the initial chain node set with directed edges according to the business flow direction, and label the business rule constraints between the nodes to obtain a chain constraint map;

[0123] Detect conflicts in the business rules between the nodes based on the chain constraint map, and use a graph traversal algorithm to verify the integrity of the constraint relationships between the nodes to obtain verification result data;

[0124] Optimize and adjust the abnormal nodes in the chain constraint map according to the verification result data, recalculate the constraint relationships between the nodes to obtain a target chain map, and establish an association index between the target chain map and the business rule library to generate a business relationship chain.

[0125] Specifically, extract key information from the optimal supplier sequence, including supplier numbers, customer numbers, and relevant business rules, and perform a combined operation using timestamps and random strings to ensure the uniqueness of each business chain. Assuming the timestamp is T1 and the random string is R1, the chain number L1 is obtained through a hash operation:

[0126] L1 = HASH(T1 || R1);

[0127] Where || represents string concatenation, and HASH(·) uses an encryption hash function such as SHA-256 to ensure uniqueness and security. This chain number will serve as the unique identifier for the business chain for subsequent querying and management. After generating the chain number, node processing is performed on the supplier information, customer information, and business rule information in the optimal supplier sequence to form an initial chain node set. Each supplier, customer, and business rule is regarded as an independent node and assigned corresponding node attributes. Assuming the attributes of the supplier node include supplier number S1, supplier credit rating S2, and available product numbers S3, the attributes of the customer node include customer number C1, order number C2, and payment method C3, and the attributes of the business rule node include payment cycle R1, delivery requirements R2, and compliance certification R3, the initial chain node set N is expressed as:

[0128] N = {(S1, S2, S3), (C1, C2, C3), (R1, R2, R3)};

[0129] Where each node stores the corresponding attribute information for subsequent calculation and verification. After constructing the initial chain node set, directed edges are established according to the business flow to form a chain constraint graph. Each supplier node S i needs to be connected to its associated customer node C j , and is constrained by the business rule node R k . For example, if supplier S i provides product P j to customer C m , and this transaction is constrained by business rule R k , then a directed edge E is established:

[0130] E = {(S i →R k ), (R k →C j )};

[0131] Each edge not only represents the business flow but also contains specific transaction constraints such as payment methods, delivery times, quality standards, etc., forming a complete chain constraint graph. Conflict detection is performed on the business rules between nodes based on the chain constraint graph to ensure the integrity and consistency of all transaction rules. The graph traversal algorithm (such as breadth-first search BFS or depth-first search DFS) is used to verify the graph to check for rule conflicts or loop dependencies. For example, if a business rule requires the delivery time D1 not to exceed 10 days while the average delivery time D2 of the supplier exceeds 10 days, there is a rule conflict:

[0132]

[0133] Among them, when Conflict = 1, it indicates that there is a constraint conflict and adjustment is required. Check for circular dependencies in the supply chain, that is, if a supplier S i depends on supplier S j , and S j depends on S i again, it will cause supply chain blockage and optimization is required. After conflict detection is completed, verification result data is generated, and based on this, abnormal nodes are optimized and adjusted. If the delivery time of a supplier does not match the business rules, recalculate the optimal supplier or adjust the business rules to adapt to market changes. For example, if a supplier S i cannot meet the delivery requirements, reselect supplier S j , and adjust the business constraints:

[0134]

[0135] Among them, D3 represents the delivery time of the new supplier, D4 is the delivery time required by the customer, select the supplier S new closest to the customer's requirements as an alternative to optimize the chain structure. Through this adjustment, the system can dynamically optimize the business chain and ensure that all transaction rules are executed. After the adjustment is completed, the target chain graph is associated with the business rule library to form the final business relationship chain. The unique number L1 of each business chain is index-associated with the relevant record R k in the business rule library:

[0136] Index = {L1 → R k};

[0137] In the future, when updating business rules, the system automatically adjusts the relevant supplier relationships and recalculates the optimal supply chain plan to maintain the stability and optimization of the business chain.

[0138] Specifically, calculating and verifying the business constraint relationships between nodes includes: dividing the chain node graph into multiple sub-graph structures, assigning optimization agents to each sub-graph, constructing a distributed constraint optimization network to obtain a multi-agent constraint system; setting local constraint functions based on the multi-agent constraint system, transforming the business rule constraints between nodes into optimization variables to obtain a set of agent constraint functions; constructing a global coupling matrix for the set of agent constraint functions, establishing a data dependency relationship model between sub-graphs to obtain a graph network constraint model; using a gradient iteration algorithm to perform local constraint solving for each agent in the graph network constraint model to obtain an initial constraint result; optimizing and accelerating the initial constraint result, quantitatively calculating the degree of constraint violation to obtain convergent constraint data; inputting the convergent constraint data into a consistency check module to verify the compliance of the constraint results of each sub-graph with the global business rules to obtain a constraint verification result; constructing a communication fault tolerance mechanism to monitor the data transmission status between sub-graphs, activating an alternative path when a communication interruption is detected to obtain a fault tolerance processing plan; correcting the constraint verification result according to the fault tolerance processing plan to ensure the stability of the constraint relationship between nodes to obtain verification result data.

[0139] Among them, detecting conflicts in the business rules between nodes based on the chain constraint graph, using a graph traversal algorithm to verify the integrity of the constraint relationships between nodes to obtain verification result data, including: classifying the state of the node relationships in the chain constraint graph, identifying the strong, weak, and potential association relationships between nodes to obtain a node relationship state matrix; dividing the business chain into equilibrium states according to the node relationship state matrix, establishing five types of equilibrium models to obtain a chain equilibrium state distribution map; calculating the business capacity threshold between nodes based on the chain equilibrium state distribution map and performing business traffic prediction to obtain a capacity balance index; dynamically monitoring the capacity balance index, setting a capacity warning threshold, and establishing a capacity warning mechanism to obtain a capacity warning rule; analyzing the load situation of the business chain in real time according to the capacity warning rule, generating a load balancing report to obtain load distribution data; performing pattern recognition on the load distribution data through a deep learning algorithm, extracting abnormal load features to obtain an abnormal pattern library; establishing a node conflict prediction model based on the abnormal pattern library to warn of potential node conflicts to obtain a conflict prediction result; comparing and analyzing the conflict prediction result with historical verification data to establish a verification evaluation model to obtain verification result data.

[0140] In a specific embodiment, the process of executing step S4 may specifically include the following steps:

[0141] Extract customer information, supplier information, product information, and delivery requirement information from the business relationship chain, and perform format conversion according to the order field mapping rule to obtain initial order data;

[0142] Perform a digital signature on the initial order data using the SHA-256 algorithm, and encrypt the sensitive fields in the order data using the AES-256 algorithm to obtain encrypted order data;

[0143] Select a communication protocol and data format from the protocol library according to the system type of the order recipient, and perform protocol encapsulation on the encrypted order data to obtain standard order format data;

[0144] Input the standard order format data into a thread pool for concurrent task allocation, establish independent communication channels for the order data of different recipients, and obtain a concurrent processing queue;

[0145] Monitor the sending status of the orders in the concurrent processing queue in real time, record the sending time, receipt confirmation, and error information to obtain order synchronization status data, and execute a retry policy for the orders that failed to be sent based on the order synchronization status data, trigger a manual intervention process, and obtain the order synchronization result.

[0146] Specifically, extract customer information, supplier information, product information, and delivery requirement information from the business relationship chain, and perform format conversion according to the order field mapping rules to ensure that different business systems can correctly parse this data. Assume that the customer information in the business relationship chain includes customer number C1, customer name C2, payment method C3, etc., while the supplier information includes supplier number S1, supplier name S2, supplier credit rating S3, etc., the product information includes product number P1, product name P2, ordered quantity P3, unit price P4, etc., and the delivery requirement information includes delivery address D1, estimated delivery date D2, and delivery method D3. These data need to be mapped according to the standard order format fields, that is:

[0147] O i = F(C i , S i , P i , D i );

[0148] Among them, O i is the converted order field, F is the conversion function of the field mapping rule, ensuring that the data structure meets the format requirements of the order management system to form the initial order data. Encrypt the order data. Perform a digital signature on the order data using the SHA-256 algorithm, and calculate the hash value H of the order:

[0149] H = SHA 256(O);

[0150] Among them, O represents order data, and H is the calculated hash value with a fixed length, ensuring that the data is not tampered with during transmission. To protect sensitive information in the order data (such as customer payment methods, supplier credit ratings, delivery addresses, etc.), the AES-256 algorithm is used to encrypt these fields. Let the set of order sensitive fields be F s , and the key be K. Then the encryption process is as follows:

[0151] E s = AES256(F s , K);

[0152] Among them, E s is the encrypted sensitive data. The key K is managed by the secure storage system to ensure that the data is decrypted only in an authorized environment, and finally forms the encrypted order data. Select the appropriate communication protocol and data format from the protocol library according to the system type of the order recipient, and perform protocol encapsulation on the encrypted order data to obtain the standard order format data. Assume that some systems use RESTful APIs for data exchange, while some ERP systems use EDI or XML formats. Then the system needs to be adapted according to the protocol type P t of the recipient:

[0153] O f = G(E s , P t );

[0154] Among them, O f is the encapsulated order data, and G is the protocol encapsulation function. This process ensures that the order data can be correctly parsed and processed between different systems, and generates the standard order format that meets the requirements of the target system. Input the standard order format data into the thread pool for concurrent task allocation, and establish independent communication channels for the order data of different recipients to form a concurrent processing queue. Assume that the system contains N recipients, and the communication channel for each recipient is T i , then the computational model for concurrent task allocation is:

[0155]

[0156] Among them, Q represents the concurrent task queue, and f is the task scheduling function, ensuring that the order data of each recipient is transmitted through the optimal communication path and avoiding blocking of the communication channel. During the transmission of the order data, monitor the sending status of the orders in the concurrent processing queue in real time, record the sending time, received confirmation information, and error logs to ensure that all orders can be correctly processed. Let the order sending time be T s , the received confirmation status be A, and the error log be E. Then the order synchronization status data is represented as:

[0157] S o=(T s , A, E);

[0158] Among them, S o records the transmission status of each order and provides the basis for analyzing abnormal situations. If it is found that some orders fail to be sent, a retry policy needs to be executed, that is, the retry interval T r is calculated according to the exponential backoff algorithm:

[0159] T r = T0 × 2 n ;

[0160] Among them, T0 is the initial retry interval, n is the current retry count. When T r reaches the maximum allowed retry time, the system will trigger a manual intervention process. The operator manually checks the order data and performs manual order supplementation or adjusts the communication strategy to finally obtain the order synchronization result, ensuring that all orders can be successfully transmitted and enter the subsequent order processing link.

[0161] The method for automatically arranging the business relationship chain in the embodiments of the present invention has been described above. Next, the device for automatically arranging the business relationship chain in the embodiments of the present invention will be described. Please refer to Figure 2 One embodiment of the device for automatically arranging the business relationship chain in the embodiments of the present invention includes:

[0162] A standardization processing module, configured to build a business rule library based on supplier information, customer information, and business information, and perform standardization processing on customer demand data to obtain standardized business demand information;

[0163] A matching operation module, configured to perform multi-dimensional matching operations on the standardized business demand information based on the business rule library to obtain an initial supplier set, and perform dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence;

[0164] A verification module, configured to build a chain node map based on the optimal supplier sequence, calculate and verify the business constraint relationship between nodes, and generate a business relationship chain;

[0165] An output module, configured to convert the business relationship chain into a standard order format, and complete order data synchronization and output the order synchronization result through an encryption signature and a concurrent distribution mechanism.

[0166] Through the collaborative cooperation of the above-mentioned various components, by establishing a multi-dimensional rule database and a standardized data processing mechanism, the management of business rules has become more systematic and standardized; by adopting multi-dimensional matching operations and dynamic weight adjustment strategies, the intelligent and precise screening of suppliers has been achieved; by introducing chain node graph technology, the visualization management and automatic verification of business constraint relationships have been carried out; by adopting encryption signatures and concurrent distribution mechanisms, the security and efficiency of order data synchronization have been ensured; the system replaces manual operations with automated processing, reduces operation risks, improves the accuracy and efficiency of business relationship chain orchestration, and at the same time supports the dynamic adjustment and real-time response of business rules, enhancing the adaptability and scalability of the system.

[0167] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0168] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0169] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, system, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0170] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0171] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for automatically arranging a business relationship chain, characterized in that: The method comprises: Build a business rule library based on supplier information, customer information, and business information, and standardize customer demand data to obtain standardized business demand information; Based on the business rule library, a multi-dimensional matching operation is performed on the standardized business demand information to obtain an initial supplier set, and a dynamic weight adjustment is performed on the initial supplier set to obtain an optimal supplier sequence; Constructing a chain node graph according to the optimal supplier sequence, calculating and verifying the business constraint relationship between nodes, and generating a business relationship chain; The business relationship chain is converted into a standard order format, and order data synchronization is completed and order synchronization results are output through encryption signature and concurrent distribution mechanism.

2. The method for automatically arranging a business relationship chain according to claim 1, characterized in that: The business rule library is constructed based on supplier information, customer information, and business information, and customer demand data is standardized to obtain standardized business demand information, including: The supplier basic information, qualification information, product type information, production capacity information and rating information in the supplier information are classified and stored to obtain a supplier rule table; The basic customer information, credit rating information, payment terms information, return rate information and cooperation years information in the customer information are classified and stored to obtain a customer rule table; The delivery time information, quality standard information, price range information, minimum order quantity information and inventory warning information in the business information are classified and stored to obtain a business rule table; Normalizing the data in the supplier rule table, the customer rule table, and the business rule table to obtain standardized data; Based on the standardized data, a cross-table field index relationship is established, the correlation strength between the fields is calculated to obtain a correlation index table, and a rule update trigger is set according to the correlation index table to verify and update the consistency of the rule data to obtain a business rule library; Collect customer demand data and perform standardized processing to obtain standardized business demand information.

3. The method for automatically arranging a business relationship chain according to claim 2, characterized in that: The collecting of customer demand data and standardization processing to obtain standardized business demand information includes: The Web acquisition unit uses an event listener to capture the demand data input by the customer on the interface in real time to obtain the interface input data; The API interface unit uses the RESTful architecture combined with the JSON Web Token authentication mechanism to verify the demand data pushed by the external system and obtain the external demand data; Duplicate data is identified on the interface input data and the external demand data to obtain deduplicated demand data, and missing fields in the deduplicated demand data are intelligently filled to obtain customer demand data; The customer demand data is subjected to field mapping conversion according to internal data format specifications to obtain demand data in a unified format, and a processing priority is set according to product type information and delivery requirement information in the demand data in the unified format to obtain standardized business demand information.

4. The method for automatically arranging a business relationship chain according to claim 1, characterized in that: The method of performing a multi-dimensional matching operation on the standardized business demand information based on the business rule library to obtain an initial supplier set, and dynamically adjusting the weight of the initial supplier set to obtain an optimal supplier sequence includes: Filtering suppliers with supply capabilities from the business rule library according to the product type information in the standardized business demand information to obtain a product matching supplier list; Based on the customer rule table in the business rule library, the product matching supplier list is filtered by blacklist and whitelist to obtain a list of cooperative qualified suppliers; Verify the delivery capability of the list of cooperative qualified suppliers according to the delivery time requirements in the standardized business demand information to obtain a list of qualified suppliers; Calculating the capacity matching degree of the list of qualified suppliers for delivery according to the order size requirements in the standardized business demand information to obtain a list of qualified suppliers; Scoring the reputation of suppliers in the capacity matching supplier list according to historical cooperation records to obtain a reputation score list, and performing supplier capacity balance calculation based on the reputation score list and the current order volume of the supplier to obtain an initial supplier set; A comprehensive score calculation and dynamic weight adjustment are performed on the initial supplier set to obtain an optimal supplier sequence.

5. The method for automatically arranging a business relationship chain according to claim 4, characterized in that: Before the comprehensive scoring calculation and dynamic weight adjustment of the initial supplier set are performed to obtain the optimal supplier sequence, the method further includes: Building a supplier state space model based on the initial supplier set, vectorizing the supplier's production capacity state, quality state, and price state to obtain a supplier state vector; Performing action space modeling on the supplier state vector and the business goal, setting price adjustment, capacity allocation, and quality control as action dimensions, and obtaining an action space model; Perform feature extraction and dimensionality reduction processing on the supplier state vector to obtain a state feature matrix, and input the state feature matrix into the value network for iterative calculation, evaluate the action value based on historical business data, and obtain an action value matrix; Sampling and simulating the supplier chain based on the action value matrix to obtain a strategy probability distribution; An exploratory search is performed on suppliers according to the probability distribution of the strategy to obtain search results, and the search results are fed back to the value network for parameter update to obtain an optimized action strategy; Using the optimized action strategy to predict the supplier state transition, generate a state transition sequence, and calculate the cumulative reward value based on a preset reward function to obtain the optimal state path; The initial supplier set is reordered and screened according to the optimal state path to generate a candidate supplier sequence as input data for comprehensive score calculation.

6. The method for automatically arranging a business relationship chain according to claim 5, characterized in that: The step of performing comprehensive scoring calculation and dynamic weight adjustment on the initial supplier set to obtain an optimal supplier sequence includes: Obtaining a candidate supplier sequence calculated from the initial supplier set, and collecting multi-dimensional data on suppliers in the candidate supplier sequence based on price competitiveness, quality stability, and delivery timeliness to obtain a supplier evaluation data set; According to the evaluation rule table in the business rule library, weights are assigned to the supplier evaluation data set to obtain an initial weight coefficient; Calculating the importance of indicators for the historical performance data in the supplier evaluation data set to obtain an indicator weight matrix, and dynamically adjusting the initial weight coefficient based on the indicator weight matrix to obtain an optimized weight coefficient; The supplier evaluation data set is weighted averaged according to the optimized weight coefficient to obtain a comprehensive supplier score, and the suppliers are sorted in descending order based on the comprehensive supplier score to obtain an optimal supplier sequence.

7. The method for automatically arranging a business relationship chain according to claim 1, characterized in that: The step of constructing a chain node graph according to the optimal supplier sequence, calculating and verifying the business constraint relationship between nodes, and generating a business relationship chain includes: Generate a unique chain identification code based on the optimal supplier sequence, and combine the timestamp with the random string to obtain a chain number; Performing node processing on the supplier information, customer information and business rule information in the optimal supplier sequence, assigning a node identifier and a node attribute to each node, and obtaining an initial node set of the chain; Connecting the nodes in the initial node set of the chain with directed edges according to the business flow direction, marking the business rule constraints between the nodes, and obtaining a chain constraint graph; Based on the chain constraint graph, the business rules between nodes are conflict detected, and the integrity of the constraint relationship between nodes is verified by using a graph traversal algorithm to obtain verification result data; According to the verification result data, the abnormal nodes in the chain constraint graph are optimized and adjusted, the constraint relationship between the nodes is recalculated to obtain the target chain graph, and the target chain graph is associated with the business rule library to generate a business relationship chain.

8. The method for automatically arranging a business relationship chain according to claim 1, characterized in that: The converting of the business relationship chain into a standard order format, completing order data synchronization and outputting order synchronization results through encryption signature and concurrent distribution mechanism, includes: Extracting customer information, supplier information, product information, and delivery requirement information from the business relationship chain, converting the format according to the order field mapping rules, and obtaining initial order data; The initial order data is digitally signed using the SHA-256 algorithm, and sensitive fields in the order data are encrypted using the AES-256 algorithm to obtain encrypted order data; Select a communication protocol and a data format from a protocol library according to the system type of the order recipient, perform protocol encapsulation on the encrypted order data, and obtain standard order format data; Input the standard order format data into the thread pool for concurrent task allocation, establish independent communication channels for order data of different recipients, and obtain a concurrent processing queue; The order sending status in the concurrent processing queue is monitored in real time, the sending time, receiving confirmation and error information are recorded, the order synchronization status data is obtained, and a retry strategy is executed for the orders that failed to send according to the order synchronization status data, triggering the manual intervention process to obtain the order synchronization result.

9. A device for automatically arranging a business relationship chain, characterized in that: A method for automatically arranging a business relationship chain according to any one of claims 1 to 8, wherein the device for automatically arranging a business relationship chain comprises: The standardization processing module is used to build a business rule library based on supplier information, customer information, and business information, and to standardize customer demand data to obtain standardized business demand information; A matching operation module, used for performing a multi-dimensional matching operation on the standardized business demand information based on the business rule library to obtain an initial supplier set, and dynamically adjusting the weight of the initial supplier set to obtain an optimal supplier sequence; A verification module, used to construct a chain node map according to the optimal supplier sequence, calculate and verify the business constraint relationship between nodes, and generate a business relationship chain; The output module is used to convert the business relationship chain into a standard order format, complete order data synchronization and output order synchronization results through encryption signature and concurrent distribution mechanism.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is enabled to execute the method for automatically arranging a business relationship chain according to any one of claims 1 to 8.

Images