An all-industry link platform supply chain collaborative management method and system

Abstract

The application discloses a whole industry link platform supply chain collaborative management method and system, wherein the whole industry link platform supply chain collaborative management method comprises the following steps: a distributed supply chain data warehouse is constructed by using a block chain and lightweight encryption; a multi-dimensional matching degree score table is generated to recommend a supplier; an encrypted bidding file is decrypted and stored, and a comprehensive score is calculated according to quantitative indexes and expert qualitative indexes; a hierarchical prediction model is constructed to output a three-dimensional prediction result; a supply-demand gap is calculated according to a supply-demand elasticity coefficient coupling algorithm, an order is automatically generated and is synchronously transmitted to each end in real time; a supply chain digital twin model is built, an optimal transportation route is dynamically planned through a smart traffic algorithm; early warning information is automatically sent to generate a response strategy, and whole-link operation data is integrated to realize visual display; and the application aims to solve the problems of existing supply chain data fragmentation, low-efficiency supplier matching, low demand prediction accuracy, rigid logistics planning, lagging risk monitoring and poor collaboration in each link.

Description

Technical Field

[0001] This invention relates to the field of supply chain management technology, and in particular to a method and system for collaborative management of supply chain on a full-industry linkage platform. Background Technology

[0002] Current supply chain management faces core pain points such as fragmented data across multiple stages and low collaborative efficiency: supplier selection relies on manual evaluation, which is prone to matching biases due to information asymmetry; document transmission security is insufficient in the bidding and evaluation process, and scoring standards are difficult to standardize; demand forecasting is mostly based on single-dimensional data, and its accuracy is greatly affected by market fluctuations; logistics planning lacks dynamic adaptability, and changes in road conditions can easily cause transportation delays; end-to-end risk monitoring is lagging, often resulting in reactive responses after problems occur. The lack of a unified management system for data across stages and the time lag in information synchronization lead to slow overall supply chain response speed, high operating costs, and an inability to meet the demands of the entire industry chain for efficient collaboration and flexible scheduling. This invention aims to solve the problems of fragmented data, inefficient supplier matching, non-standardized bidding and evaluation, low accuracy in demand forecasting, rigid logistics planning, lagging risk monitoring, and poor collaboration and slow response across stages in the existing supply chain. Summary of the Invention

[0003] This invention provides a method for collaborative management of the supply chain on a full-industry-linked platform, comprising:

[0004] Real-time collection of multi-source data across the entire supply chain; and the construction of a distributed supply chain data warehouse using blockchain and lightweight encryption.

[0005] Based on the dual-core data, feature mapping and weight calculation are performed to generate a multi-dimensional matching score table to recommend suppliers.

[0006] Receive encrypted bid documents, decrypt and store them, calculate a comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report;

[0007] A hierarchical prediction model is built by calling multi-dimensional correlated data, and the three-dimensional prediction results are output.

[0008] Combining the three-dimensional demand forecast results, the supply-demand gap is calculated based on the supply-demand elasticity coefficient coupling algorithm, and orders are automatically generated and their status is synchronized to each end in real time.

[0009] Build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route through the intelligent traffic algorithm;

[0010] Real-time monitoring of risk factors; when a risk signal is detected, automatic early warning information is sent and response strategies are generated; and full-chain operational data is integrated and visualized.

[0011] The supply chain collaborative management method of the full-industry linkage platform described above includes real-time collection of multi-source data across the entire supply chain and the construction of a distributed supply chain data warehouse using blockchain and lightweight encryption, comprising:

[0012] Real-time collection of end-to-end data from suppliers, manufacturers, logistics providers, distributors, and end customers, followed by standardized preprocessing;

[0013] The processed data blocks are encrypted using a lightweight encryption algorithm, and the hash value and encrypted data are uploaded to the blockchain network to build a supply chain data warehouse.

[0014] The supply chain collaborative management method of the full-industry linkage platform described above includes the generation of a multi-dimensional matching score table to recommend suppliers based on feature mapping and weight calculation using dual-core data, including:

[0015] Based on the demand characteristics of the purchaser and the capability characteristics of potential suppliers, a feature mapping model is established to vectorize the features of both the supply and demand sides and place them in the same metric space for comparison.

[0016] Using the analytic hierarchy process (AHP), the weights of each feature dimension are dynamically calculated based on the current business scenario to generate a multi-dimensional matching degree for suppliers, thus forming a supplier recommendation list.

[0017] The supply chain collaborative management method of the full-industry linkage platform described above includes receiving encrypted bid documents, decrypting and storing them, calculating a comprehensive score based on quantitative indicators and expert qualitative indicators, and generating a structured bid evaluation report, including:

[0018] Receive digitally signed and encrypted tender documents, perform secure decryption using a pre-built key management service, and verify the validity of the digital signature. Verification is achieved by automatically parsing the structured data and storing it in a temporary database.

[0019] The system automatically calculates the scores of quantitative indicators and simultaneously calculates the final comprehensive score of each bidder based on the online qualitative indicators of the review experts according to preset weights, and automatically generates a structured bid evaluation report.

[0020] The supply chain collaborative management method of the full-industry linkage platform described above includes, in which, by combining three-dimensional demand forecasting results and calculating the supply-demand gap according to a supply-demand elasticity coefficient coupling algorithm, orders are automatically generated and their status is synchronized to each end in real time, including:

[0021] Based on the construction of a global supply and demand view, the supply and demand gap in the future time period is dynamically calculated through the supply and demand elasticity coefficient coupling algorithm.

[0022] The system automatically triggers order generation logic based on supply and demand gaps, generates purchase orders for suppliers, generates production work orders for the internal production system, and simultaneously notifies all participating parties of the order status.

[0023] The supply chain collaborative management method of the full-industry linkage platform described above includes building a digital twin model of the supply chain, retrieving real-time logistics data, and dynamically planning the optimal transportation route through a smart traffic algorithm, including:

[0024] A digital twin model of the supply chain logistics process is built based on geographic information system, warehousing and transportation network node data;

[0025] The intelligent traffic algorithm analyzes and processes real-time data streams, dynamically simulates and calculates the current optimal transportation route, and can dynamically reroute vehicles en route.

[0026] The supply chain collaborative management method of the full-industry chain platform described above includes real-time monitoring of risk factors, automatic sending of early warning information and generation of response strategies when risk signals are detected, and integration and visualization of full-chain operational data, including:

[0027] A risk rule base is set up to scan and identify real-time data across the entire data warehouse. When a risk signal is detected and reaches the warning threshold, an early warning mechanism is automatically triggered.

[0028] We continuously integrate operational data from all aspects and present a panoramic view of the supply chain's operational status in an intuitive way through a visual dashboard.

[0029] A supply chain collaborative management system for a full-industry chain platform, comprising:

[0030] The ChainSec Data Warehouse module is used to collect multi-source data from the entire supply chain in real time, and uses blockchain and lightweight encryption to build a distributed supply chain data warehouse.

[0031] The supplier recommendation module is used to perform feature mapping and weight calculation based on dual core data, and generate a multi-dimensional matching score table to recommend suppliers.

[0032] The bid evaluation report module is used to receive encrypted bid documents, decrypt and store them, calculate the comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report.

[0033] The demand forecasting module is used to call multi-dimensional correlated data to build a hierarchical forecasting model and output three-dimensional forecasting results;

[0034] The order generation module combines the three-dimensional demand forecast results, calculates the supply and demand gap based on the supply and demand elasticity coefficient coupling algorithm, automatically generates orders, and synchronizes the order status to each end in real time.

[0035] The logistics planning module is used to build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route through the intelligent traffic algorithm.

[0036] The risk monitoring module is used to monitor risk factors in real time. When a risk signal is detected, it automatically sends early warning information, generates response strategies, integrates full-chain operational data, and displays it visually.

[0037] The beneficial effects achieved by this invention are as follows:

[0038] By constructing a distributed data warehouse using blockchain encryption, secure storage and real-time sharing of data across the entire supply chain are achieved, eliminating information silos and enhancing data credibility. Intelligent algorithms enable precise supplier matching, three-dimensional demand prediction, and dynamic logistics planning, replacing traditional manual decision-making, reducing human error, and improving supply chain operational efficiency. Automated collaboration across all stages, from supplier selection and bidding evaluation to order generation and logistics scheduling, is realized, shortening response cycles, reducing operating costs, and providing stable technical support for the collaborative development of the entire industry chain. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0040] Figure 1 This is a flowchart of a supply chain collaborative management method for a full-industry linkage platform provided in Embodiment 1 of this application.

[0041] Figure 2 This is a schematic diagram of a supply chain collaborative management system for a full-industry linkage platform provided in Embodiment 2 of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1

[0044] like Figure 1 As shown, Embodiment 1 of this application provides a supply chain collaborative management method for a full-industry chain platform, including:

[0045] S1: Real-time collection of multi-source data across the entire supply chain, using blockchain and lightweight encryption to build a distributed supply chain data warehouse.

[0046] The process involves real-time collection of multi-source data across the entire supply chain and the construction of a distributed supply chain data warehouse using blockchain and lightweight encryption. This includes the following sub-steps:

[0047] S11: Real-time collection of end-to-end data from suppliers, manufacturers, logistics providers, distributors, and end customers, followed by standardized preprocessing.

[0048] The system deploys distributed data acquisition nodes, adapting to multi-source data protocols from supplier ERP systems, manufacturer MES data interfaces, logistics provider GPS tracking systems, distributor inventory management platforms, and end-customer feedback channels. This establishes a real-time data access channel for concurrent data acquisition across the entire supply chain. It identifies and removes abnormal data through pre-defined anomaly detection rules, completing data cleaning. Based on a supply chain data standard mapping library, it uniformly converts heterogeneous data into a structured format. Combining a business scenario tagging system, the rule engine automatically matches the data to its supply chain link, business type, and related objects, generating standardized data blocks containing multi-level tags.

[0049] S12: Encrypt the processed data blocks using a lightweight encryption algorithm, and upload the hash value and encrypted data to the blockchain network to build a supply chain data warehouse.

[0050] After receiving and processing the standardized data block, a dynamic key derivation mechanism is introduced on the basis of the national cryptographic SM4 algorithm. First, the core features in the data block tag are extracted, and a special key bound to the data features is generated through a preset key derivation function. The SM4 block encryption is performed using this key. At the same time, a dynamic factor based on timestamp is embedded in the encryption process, so that the encryption results of the same data block are different at different times.

[0051] After encryption, a double hash value is calculated on the ciphertext (the primary hash is based on the entire ciphertext, and the secondary hash is based on the segmented ciphertext). The double hash, key derivation parameters, and dynamic factor metadata are packaged into a blockchain transaction and written to the ledger after consensus among the consortium blockchain nodes. Encrypted data blocks are distributed and stored in an off-chain sharded cluster according to feature tags. The double hash index in the blockchain ledger enables fast location and integrity verification, retaining the efficiency of the SM4 algorithm, and improving anti-cracking capabilities through dynamic keys and double verification.

[0052] S2: Based on the dual-core data, perform feature mapping and weight calculation to generate a multi-dimensional matching score table to recommend suppliers.

[0053] The process of performing feature mapping and weight calculation based on dual-core data to generate a multi-dimensional matching score table for supplier recommendations includes the following sub-steps:

[0054] S21: Based on the demand characteristics of the purchaser and the capability characteristics of potential suppliers, establish a feature mapping model, vectorize the features of both the supply and demand sides, and place them in the same metric space for comparison.

[0055] The system extracts buyer demand characteristics and potential supplier capability characteristics from the supply chain data warehouse. A feature parsing engine then structures these two types of data, identifying core indicators in the demand (such as delivery cycle, quality standards, and cost thresholds) and supplier capability dimensions (such as production capacity, certifications, and historical performance records). Based on the supply chain business scenario, the system automatically matches the correspondence between supply and demand characteristics, establishing a multi-dimensional mapping matrix. A spatial mapping algorithm projects the feature vectors of both supply and demand sides onto a pre-defined high-dimensional space. Coordinate calibration eliminates differences in feature dimensions, generating a standardized feature vector set.

[0056] The feature parsing engine uses a knowledge graph in the supply chain domain as its underlying architecture. The graph predefines core entities (such as delivery and quality) and attributes and business rules related to procurement needs and supplier capabilities. It also embeds a dynamic rule learning submodule, which is continuously trained through historical parsing cases and real-time business feedback data to automatically adjust feature recognition thresholds and parsing logic. Structured data is directly matched with graph fields to complete extraction, while unstructured data (such as demand description text and qualification certificates) is mapped to graph nodes through semantic segmentation and entity linking technology to dynamically generate a set of core indicators and capability dimension labels with confidence.

[0057] S22: Using the analytic hierarchy process (AHP), the weights of each feature dimension are dynamically calculated based on the current business scenario to generate a multi-dimensional matching degree for suppliers and form a supplier recommendation list.

[0058] Based on a standardized feature vector set, a dynamic hierarchical analysis model is constructed based on business scenarios, dividing the feature dimensions into target, criterion, and solution layers. The scenario feature extraction module parses current business attributes to generate a scenario factor matrix, dynamically corrects the criterion layer judgment matrix, and adjusts element values ​​by combining historical case feature influence statistics. The weights of each dimension are calculated using the eigenvalue method and pass a consistency check. Subsequently, a multi-dimensional matching degree calculation formula is introduced:

[0059] and These are the start time and the end time, respectively; For the index of the feature dimension, This represents the total number of feature dimensions. Indicates the first Dynamic weighting function for each feature dimension For scene parameters and It also satisfies that the sum of the weights of all dimensions is 1, and its value is dynamically adjusted according to changes in business scenarios to reflect the importance of each feature dimension under different scenarios. Indicates that the supplier in the Temporal vectors in each feature dimension This is a time variable, reflecting the dynamic changes of this feature dimension over time. Indicates the first The historical influence coefficients of each feature dimension are updated in real time using a Bayesian estimation algorithm, reflecting the weight of the influence of the historical performance of that feature dimension on the current matching degree. Represents the exponentially decaying term, where yes The first derivative with respect to time (i.e., the characteristic rate of change); This represents the covariance matrix between feature dimensions, reflecting the correlation between different feature dimensions. Its determinant value is used to quantify the impact of the degree of coupling between feature dimensions on the overall matching degree. This represents the Laplacian operator, which enhances the impact of variations in the spatial distribution of features on the matching degree. This represents the feature matching deviation matrix, where the elements reflect the degree of deviation between the actual and ideal values ​​of each feature dimension. This indicates a real-time adjustment factor vector, whose elements are dynamically generated based on real-time business data, used to correct the impact of feature matching bias on the final result. This represents the Hadamard product operator, used to implement element-level interactions such as feature matching bias and real-time adjustment factors. Denotes the shock response function vector, where The lag time reflects the degree of impact of external shocks on feature matching under different lag times. This represents the real-time perturbation vector. This represents the current moment corresponding to the lag time, and the elements in the vector reflect the external disturbances that occur in real time.

[0060] Perform a weighted operation on the feature vectors of each supplier and their corresponding weights to generate multi-dimensional matching sub-scores and comprehensive scores. Sort the comprehensive scores and associate them with feature matching details to generate a supplier recommendation list.

[0061] S3: Receive encrypted bid documents, decrypt and store them, calculate the comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report.

[0062] The process of receiving encrypted bid documents, decrypting and storing them, calculating a comprehensive score based on quantitative and expert qualitative indicators, and generating a structured bid evaluation report includes the following sub-steps:

[0063] S31: Receive digitally signed and encrypted tender documents, perform secure decryption using a pre-built key management service, and verify the validity of the digital signature. Verification is achieved by automatically parsing the structured data and storing it in a temporary database.

[0064] A dedicated channel for receiving encrypted bid documents is established, enabling file transmission across different encryption formats via a protocol adaptation module. Upon receipt, the received documents are automatically associated with the bidder's identity and temporarily stored in an isolated buffer. A pre-built key management service is then invoked, employing a layered decryption mechanism of "master key + session key." First, the session key is unlocked via a hardware encryption module, and then used to decrypt the bid document. After decryption, the digital signature information in the file is extracted and compared with the bidder's public key pre-stored in the supply chain data warehouse. Simultaneously, a hash value verification algorithm verifies the file's integrity. Once both verifications pass, a structured parsing engine is activated, automatically extracting core data such as price, qualifications, and performance commitments according to the standard field templates in the bid document. This data is then mapped to corresponding fields in a temporary database and indexed.

[0065] S32: Automatically calculates the scores of quantitative indicators, and at the same time calculates the final comprehensive score of each bidder based on the online qualitative indicators of the review experts according to the preset weights, and automatically generates a structured bid evaluation report.

[0066] Retrieve the parsed bidding data, first start the quantitative indicator calculation engine, extract quantifiable data such as price and delivery cycle according to preset standards, and then use formulas.

[0067] For indexing quantitative indicators, The total number of quantitative indicators; Indicates the first The weighting coefficients of the quantitative indicators reflect the importance of different quantitative indicators in the comprehensive evaluation. Indicates the first Real-time monitoring values ​​of quantitative indicators, It refers to time-varying quantifiable data, such as price and delivery cycle, which change over time. Indicates the first The benchmark threshold for each indicator is generated by fitting industry standards with historical best data, and serves as a reference benchmark for measuring indicator performance. Indicates the first The calibration sensitivity coefficient of each indicator is dynamically adjusted according to the business scenario. The larger the value, the higher the sensitivity to deviations from the indicator. Indicates the first Items in time The cumulative effect of abnormal deviations within the range is as follows: , Indicates the first Item In Abnormal deviation at any given time. Represents the exponentially decaying term. This is the abnormal decay time constant, used to simulate the decay trend of abnormal effects over time. Indicates the first The abnormal decay time constant of the indicator reflects the duration of the abnormal impact. Indicates the first The first derivative of the abnormal deviation of an indicator with respect to time reflects the rate of change of the abnormal deviation and is used to capture the dynamic characteristics of abnormal trends.

[0068] The system automatically calculates quantitative scores for each bidder, correcting for outlier data using a dynamic threshold calibration module before calculation. Subsequently, an online expert review portal is opened to collect expert scores on qualitative indicators such as qualification compliance and feasibility of the proposal. After outlier scores are removed using a consistency check algorithm, the final score is determined by combining multi-dimensional matching criteria.

[0069] Indicates the multi-dimensional matching score; Scores for quantitative indicators; Q represents the expert weighting coefficient; Q is the mean of the expert qualitative scores after consistency verification. This represents the scenario adjustment factor, used to dynamically adjust the overall score to suit different business scenarios. Its value is set according to the specific business attributes. The quantitative indicator score F represents the time... The first derivative of reflects the rate of change of the quantitative index score over time; The S-score represents the multi-dimensional matching degree versus time. The first derivative reflects the rate of change of the multidimensional matching degree over time.

[0070] Calculate the overall score of suppliers, and finally sort and integrate the scoring details according to the score to automatically generate a structured bid evaluation report containing the index calculation process and expert review opinions.

[0071] S4: Build a hierarchical prediction model by calling multi-dimensional correlated data and output three-dimensional prediction results.

[0072] By calling upon multi-dimensional correlated data, a hierarchical prediction model is constructed to output refined demand forecasts for different regions and products within a specific future period.

[0073] By leveraging multi-dimensional correlated data from the supply chain data warehouse, including historical sales, inventory turnover, and regional market dynamics, a hierarchical predictive model is constructed. The bottom layer uses time series algorithms to capture basic demand trends for single products; the middle layer uses regional feature clustering algorithms to group and analyze regions with similar consumption patterns; and the top layer incorporates external variables such as promotional activities using a market influencing factor model.

[0074] The model parameters are optimized through rolling iterations. The final output is a refined demand forecast that integrates three dimensions: a specific future period, different regions, and different products. Each dimension's forecast value is accompanied by a confidence interval.

[0075] S5: Combining the three-dimensional demand forecast results, the supply and demand gap is calculated based on the supply and demand elasticity coefficient coupling algorithm, and orders are automatically generated and their status is synchronized to each end in real time.

[0076] The process involves combining the three-dimensional demand forecast results, calculating the supply-demand gap using a supply-demand elasticity coefficient coupling algorithm, automatically generating orders, and synchronizing the order status to each end in real time. This includes the following sub-steps:

[0077] S51: Based on the construction of a global supply and demand view, the supply and demand gap in the future time period is dynamically calculated through the supply and demand elasticity coefficient coupling algorithm.

[0078] Based on the three-dimensional demand forecast results, internal capacity data and real-time inventory data at all levels are simultaneously retrieved to construct a three-dimensional global supply and demand view. Data cube technology is used to achieve dynamic correlation and visualization mapping of multi-dimensional data. On this basis, a supply and demand elasticity coefficient coupling algorithm is employed to dynamically calculate the supply and demand gap, simultaneously outputting the gap fluctuation range and influencing factor analysis.

[0079] The formula for calculating the supply-demand gap is:

[0080] This indicates a real-time feedback correction factor; This represents a dynamic coefficient, serving as a coupling bridge between demand elasticity and supply elasticity. When the market is dominated by demand fluctuations, its value approaches 1; when supply constraints are more significant, its value approaches 0, achieving scenario-based adaptation of supply and demand elasticity. This represents the three-dimensional demand forecast value at time t, which is the total future demand predicted by multiple factors. This represents the market volatility coefficient at time t, reflecting the intensity of market volatility on the demand side. Indicates the current point in time in the calculation; express The policy adjustment coefficient at any given time reflects the degree to which changes in demand-side policies affect demand. It is the first derivative of the demand forecast with respect to time, reflecting the instantaneous rate of change of demand over time; Indicates the corresponding Three-dimensional demand forecast values ​​at any given time; It is a sine function of the market volatility coefficient, simulating the periodic characteristics of market fluctuations; This represents the overall supply capacity value at time t; This represents the production elasticity coefficient at time t; Indicates the time response threshold; It is the time response threshold at time s, reflecting the speed at which the supply side responds to changes in demand.

[0081] The supply-demand elasticity coefficient coupling algorithm is based on the dynamic correlation between supply and demand and is designed in three progressive layers: The basic layer identifies the core variables affecting supply and demand by analyzing historical data (focusing on market fluctuations and policy adjustments on the demand side, and on the supply side, focusing on capacity elasticity and time response thresholds), and constructs calculation models for demand elasticity and supply elasticity respectively to ensure that the single-dimensional elasticity can accurately reflect the sensitivity of each variable; The correlation layer designs a coupling mechanism and introduces a dynamic coefficient as a bridge. When demand fluctuations dominate the market, the dynamic coefficient approaches 1, and when supply constraints are more significant, the dynamic coefficient approaches 0, realizing the scenario-based adaptation of supply and demand elasticity; The optimization layer embeds a real-time feedback correction module, which inputs the actual supply and demand deviation data back into the algorithm. The weighting rules of dynamic parameters and dynamic coefficients ensure that the coupling result can not only reflect the elasticity characteristics of supply and demand, but also capture the implicit correlation between them (such as the inhibitory effect of demand surge on supply elasticity). Finally, the output is a comprehensive supply and demand gap that takes into account both the immediate gap and trend changes.

[0082] S52: Automatically triggers order generation logic based on supply and demand gaps, generates purchase orders for suppliers, generates production work orders for the internal production system, and simultaneously notifies all participants of the order status.

[0083] Based on supply and demand gap data, the order generation engine is activated. First, it matches the gap size with product type, regional distribution, and business priority to pre-defined order triggering rules. When the gap meets the triggering conditions, the order generation process is automatically activated. The system calls upon the supplier recommendation list output by S2, selecting the best suppliers based on comprehensive scores. Simultaneously, it connects with the internal production system to assess existing capacity and production cycle, dynamically determining the order allocation ratio between external procurement and internal production. Purchase orders are pushed to the selected suppliers, and production work orders containing process arrangements and completion deadlines are issued to the internal production system. All orders are assigned a unique blockchain identifier. Through a distributed message synchronization mechanism, the system pushes order status to all participants in the supply chain in real time and writes order flow data to the blockchain for notarization, ensuring that order information is traceable and tamper-proof.

[0084] S6: Build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route through the intelligent traffic algorithm.

[0085] The process of building a digital twin model of the supply chain and dynamically planning the optimal transportation route using the intelligent traffic algorithm, includes the following sub-steps:

[0086] S61: Based on geographic information system, warehousing and transportation network node data, build a digital twin model of the supply chain logistics links.

[0087] By combining road network data from a geographic information system, spatial distribution data of warehouse nodes, and infrastructure parameters of the transportation network, a foundational data pool for a supply chain digital twin is constructed. A spatiotemporal coordinate alignment algorithm maps multi-source heterogeneous data to a unified spatial coordinate system. A layered modeling architecture is adopted: the physical layer performs 3D digital reconstruction of entities such as warehouses, transportation hubs, and distribution points, assigning them physical attributes; the logical layer constructs transportation path relationships between nodes through a directed graph model, embedding constraint parameters such as road grade, traffic restrictions, and historical traffic efficiency; the interaction layer develops a real-time data access interface to achieve seamless integration with dynamic data sources such as vehicle GPS and warehouse sensors. The system automatically corrects spatial relationship errors between nodes and paths in the model and dynamically calibrates model parameters through deviation analysis between historical operational data and the physical world, enabling the digital twin model to accurately map the real state of the supply chain logistics network.

[0088] S62: It analyzes and processes real-time data streams through intelligent traffic condition algorithms, dynamically simulates and calculates the current optimal transportation route, and can dynamically reroute vehicles on the way.

[0089] Based on the supply chain digital twin model, a dynamic data stream is formed by connecting to the road condition monitoring system, meteorological platform, and GPS and status data of vehicles en route via real-time data interfaces. The intelligent road condition algorithm first performs spatiotemporal alignment and noise filtering on multi-source data, converting heterogeneous data such as road congestion index, weather impact coefficient, and road construction information into unified road segment cost parameters. The weights of various influencing factors are adjusted in real-time according to the timeliness requirements of the transportation task and the attributes of the goods—increasing the weight of real-time road conditions in emergency transportation scenarios and strengthening the proportion of road carrying capacity factors for bulk cargo transportation. An improved A* path search algorithm simulates multiple candidate paths in the digital twin model, combining historical traffic efficiency data and real-time cost parameters to calculate a comprehensive path score and select the optimal transportation route. When a sudden change in road conditions is detected, a dynamic rerouting mechanism is triggered, rapidly iterating and calculating a new path in the twin model and simultaneously pushing it to the terminals of vehicles en route and the dispatch center.

[0090] S7: Real-time monitoring of risk factors; when a risk signal is detected, it automatically sends early warning information, generates response strategies, integrates full-chain operational data, and displays it visually.

[0091] The process includes real-time monitoring of risk factors, automatic sending of early warning information and generation of response strategies when risk signals are detected, and generation and visualization of a comprehensive supply chain operation decision report. This includes the following sub-steps:

[0092] S71: Set up a risk rule base, scan and identify real-time data across the entire data warehouse, and automatically trigger an early warning mechanism when a risk signal is detected and reaches the early warning threshold.

[0093] Based on a logistics digital twin model, real-time traffic conditions, vehicle trajectories, and path deviation data processed by the intelligent traffic algorithm are deeply reused and incorporated into a full-chain risk monitoring system. A risk rule base covering all links of the supply chain is established. The full-chain data of the data warehouse is scanned through a real-time stream processing engine, and risk characteristics are compared using pattern recognition algorithms. The logistics risk analysis directly calls the dynamic path data in the twin model and combines it with a predictive model trained on historical delay cases to identify potential transportation disruption signals in advance. Dynamic early warning thresholds are set. When the risk signal in any link reaches the threshold, or when a cross-link chain reaction trend is detected by the risk transmission algorithm, an early warning mechanism is immediately triggered, automatically pushing tiered early warning information to relevant nodes and calling a pre-set response strategy library to generate solutions including logistics emergency routing and alternative supplier switching.

[0094] S72: Continuously integrate operational data from all aspects and present a panoramic view of the supply chain's operational status in an intuitive way through a visual dashboard.

[0095] Continuously connect with real-time operational data from all links of the supply chain, including supplier fulfillment data, production progress data, logistics and transportation data, inventory change data, and risk warning information. Based on the supply chain digital twin model, a multi-dimensional visualization engine is built. The bottom layer is the basic indicator layer, which displays core data of each link with dynamic charts; the middle layer is the correlation analysis layer, which presents the data flow and dependencies across links through Sankey diagrams and network diagrams; the top layer is the comprehensive decision-making layer, which integrates key performance indicators to form a supply chain health score. For abnormal data, visual highlighting is automatically triggered, and linked annotations are generated based on the risk warning results.

[0096] Example 2

[0097] like Figure 2 As shown, Embodiment 2 of this application provides a supply chain collaborative management system for a full-industry chain platform, including:

[0098] ChainSec Data Warehouse Module 21: Used for real-time collection of multi-source data across the entire supply chain, and employs blockchain and lightweight encryption to build a distributed supply chain data warehouse;

[0099] Supplier Recommendation Module 22: Used to perform feature mapping and weight calculation based on dual core data, and generate a multi-dimensional matching score table to recommend suppliers;

[0100] Module 23: Used to receive encrypted bid documents, decrypt and store them, calculate the comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report;

[0101] Demand Forecasting Module 24: Used to call multi-dimensional correlated data to build a hierarchical forecasting model and output three-dimensional forecasting results;

[0102] Order generation module 25: It is used to combine the three-dimensional demand forecast results, calculate the supply and demand gap according to the supply and demand elasticity coefficient coupling algorithm, automatically generate orders, and synchronize the order status to each terminal in real time;

[0103] Logistics Planning Module 26: Used to build a digital twin model of the supply chain, retrieve real-time logistics data and dynamically plan the optimal transportation route through the intelligent traffic algorithm;

[0104] Risk monitoring module 27: Used to monitor risk factors in real time. When a risk signal is detected, it automatically sends early warning information, generates response strategies, integrates full-chain operation data, and displays it visually.

[0105] Corresponding to the above embodiments, the present invention provides a computer storage medium, including: at least one memory and at least one processor;

[0106] The memory is used to store one or more program instructions;

[0107] A processor is used to run one or more program instructions to execute a supply chain collaborative management method for a full-industry-linked platform.

[0108] Corresponding to the above embodiments, this embodiment of the invention provides a computer-readable storage medium containing one or more program instructions, which are executed by a processor to provide a supply chain collaborative management method for a full-industry linkage platform.

[0109] The embodiments disclosed in this invention provide a computer-readable storage medium storing computer program instructions. When the computer program instructions are executed on a computer, the computer performs the aforementioned supply chain collaborative management method for a full-industry linkage platform.

[0110] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0111] The various methods, steps, and logic diagrams disclosed in the embodiments of this invention can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods.

[0112] The storage medium can be memory, such as volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0113] Among them, non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0114] Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM).

[0115] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0116] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in this invention can be implemented using a combination of hardware and software. When applied as software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0117] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of the present invention should be included within the scope of protection of the present invention.

Claims

1. A supply chain collaborative management method for a full-industry chain platform, characterized in that, include: Real-time collection of multi-source data across the entire supply chain; and the construction of a distributed supply chain data warehouse using blockchain and lightweight encryption. Based on the dual-core data, feature mapping and weight calculation are performed to generate a multi-dimensional matching score table to recommend suppliers. Receive encrypted bid documents, decrypt and store them, calculate a comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report; A hierarchical prediction model is built by calling multi-dimensional correlated data, and the three-dimensional prediction results are output. Combining the three-dimensional demand forecast results, the supply-demand gap is calculated using a supply-demand elasticity coefficient coupling algorithm. Orders are automatically generated and their status is synchronized to all parties in real time. The formula for calculating the supply-demand gap is as follows: , This indicates a real-time feedback correction factor; This represents a dynamic coefficient, serving as a coupling bridge between demand elasticity and supply elasticity. When the market is dominated by demand fluctuations, its value approaches 1; when supply constraints are more significant, its value approaches 0, achieving scenario-based adaptation of supply and demand elasticity. This represents the three-dimensional demand forecast value at time t, which is the total future demand predicted by multiple factors. This represents the market volatility coefficient at time t, reflecting the intensity of market volatility on the demand side. Indicates the current point in time in the calculation; The policy adjustment coefficient at time t represents the degree of impact of changes in demand-side policies on demand. It is the first derivative of the demand forecast with respect to time, reflecting the instantaneous rate of change of demand over time; Indicates the corresponding Three-dimensional demand forecast values ​​at any given time; It is a sine function of the market volatility coefficient, simulating the periodic characteristics of market fluctuations; This represents the overall supply capacity value at time t; This represents the production elasticity coefficient at time t; Indicates the time response threshold; It is the time response threshold at time s, reflecting the speed at which the supply side responds to changes in demand; Build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route through the intelligent traffic algorithm; Real-time monitoring of risk factors; when a risk signal is detected, automatic early warning information is sent and response strategies are generated; and full-chain operational data is integrated and visualized. Specifically, the use of blockchain and lightweight encryption to build a distributed supply chain data warehouse involves encrypting the processed data blocks using a lightweight encryption algorithm, and uploading the hash value and the encrypted data to the blockchain network to build the supply chain data warehouse. After receiving and processing the standardized data block, a dynamic key derivation mechanism is introduced on the basis of the national cryptographic SM4 algorithm. First, the core features in the data block tag are extracted, and a special key bound to the data features is generated through a preset key derivation function. The key is used to perform SM4 block encryption. At the same time, a dynamic factor based on timestamp is embedded in the encryption process, so that the encryption results of the same data block are different at different times. After encryption, a double hash value is calculated on the ciphertext. The main hash is based on the entire ciphertext, and the secondary hash is based on the segmented ciphertext. The double hash, key derivation parameters, and dynamic factor metadata are packaged into a blockchain transaction and written into the ledger after consensus among the consortium blockchain nodes. The encrypted data blocks are distributed and stored in an off-chain sharded cluster according to feature tags. Specifically, based on a standardized feature vector set, a dynamic hierarchical analysis model is constructed based on business scenarios, dividing the feature dimensions into target, criterion, and solution layers. The scenario feature extraction module parses current business attributes to generate a scenario factor matrix, dynamically corrects the criterion layer judgment matrix, and adjusts element values ​​by combining historical case feature influence statistical learning. The weights of each dimension are calculated using the eigenvalue method and pass a consistency check. Subsequently, a multi-dimensional matching degree calculation formula is introduced: , and These are the start time and the end time, respectively; For the index of the feature dimension, This represents the total number of feature dimensions. Indicates the first Dynamic weighting function for each feature dimension For scene parameters and It also satisfies that the sum of the weights of all dimensions is 1, and its value is dynamically adjusted according to changes in business scenarios to reflect the importance of each feature dimension in different scenarios. Indicates that the supplier in the Temporal vectors in each feature dimension As a time variable, it reflects the dynamic changes of this feature dimension over time; Indicates the first The historical influence coefficients of each feature dimension are updated in real time using a Bayesian estimation algorithm, reflecting the weight of the historical performance of that feature dimension on the current matching degree. Represents the exponentially decaying term, where yes The first derivative with respect to time, i.e., the characteristic rate of change; This represents the covariance matrix between feature dimensions, reflecting the correlation between different feature dimensions. Its determinant value is used to quantify the impact of the degree of coupling between feature dimensions on the overall matching degree. This represents the Laplacian operator, which enhances the impact of variations in the spatial distribution of features on the matching degree. This represents the feature matching deviation matrix, where the elements reflect the degree of deviation between the actual and ideal values ​​of each feature dimension. This indicates that the factor vector is adjusted in real time. The elements in the vector are dynamically generated based on real-time business data and are used to correct the impact of feature matching deviation on the final result. This represents the Hadamard product operator, used to implement element-level interactions such as feature matching bias and real-time adjustment factors. Denotes the shock response function vector, where The lag time reflects the degree of impact of external shocks on feature matching under different lag times; This represents the real-time perturbation vector. This represents the current moment corresponding to the lag time, and the elements in the vector reflect the external disturbances that occur in real time. Perform a weighted operation on the feature vectors of each supplier and their corresponding weights to generate multi-dimensional matching sub-scores and comprehensive scores. Sort the comprehensive scores and associate them with feature matching details to generate a supplier recommendation list.

2. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, Real-time collection of multi-source data across the entire supply chain; construction of a distributed supply chain data warehouse using blockchain and lightweight encryption, including: Real-time collection of end-to-end data from suppliers, manufacturers, logistics providers, distributors, and end customers, followed by standardized preprocessing; The processed data blocks are encrypted using a lightweight encryption algorithm, and the hash value and encrypted data are uploaded to the blockchain network to build a supply chain data warehouse.

3. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, Based on the dual-core data, feature mapping and weight calculation are performed to generate a multi-dimensional matching score table to recommend suppliers, including: Based on the demand characteristics of the purchaser and the capability characteristics of potential suppliers, a feature mapping model is established to vectorize the features of both the supply and demand sides and place them in the same metric space for comparison. Using the analytic hierarchy process (AHP), the weights of each feature dimension are dynamically calculated based on the current business scenario to generate a multi-dimensional matching degree for suppliers, thus forming a supplier recommendation list.

4. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, The system receives encrypted bid documents, decrypts and stores them, calculates a comprehensive score based on quantitative and expert qualitative indicators, and generates a structured bid evaluation report, including: Receive digitally signed and encrypted tender documents, perform secure decryption using a pre-built key management service, and verify the validity of the digital signature. Verification is achieved by automatically parsing the structured data and storing it in a temporary database. The system automatically calculates the scores of quantitative indicators and simultaneously calculates the final comprehensive score of each bidder based on the online qualitative indicators of the review experts according to preset weights, and automatically generates a structured bid evaluation report.

5. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, Combining the three-dimensional demand forecast results, the supply-demand gap is calculated using a supply-demand elasticity coefficient coupling algorithm. Orders are automatically generated and their status is synchronized to all terminals in real time, including: Based on the construction of a global supply and demand view, the supply and demand gap in the future time period is dynamically calculated through the supply and demand elasticity coefficient coupling algorithm. The system automatically triggers order generation logic based on supply and demand gaps, generates purchase orders for suppliers, generates production work orders for the internal production system, and simultaneously notifies all participating parties of the order status.

6. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, Build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route using intelligent traffic algorithms, including: A digital twin model of the supply chain logistics process is built based on geographic information system, warehousing and transportation network node data; The intelligent traffic algorithm analyzes and processes real-time data streams, dynamically simulates and calculates the current optimal transportation route, and can dynamically reroute vehicles en route.

7. The supply chain collaborative management method for a full-industry chain platform according to claim 1, characterized in that, Real-time monitoring of risk factors; automatic sending of early warning information and generation of response strategies when risk signals are detected; integration of end-to-end operational data and visualization, including: A risk rule base is set up to scan and identify real-time data across the entire data warehouse. When a risk signal is detected and reaches the warning threshold, an early warning mechanism is automatically triggered. We continuously integrate operational data from all aspects and present a panoramic view of the supply chain's operational status in an intuitive way through a visual dashboard.

8. A supply chain collaborative management system for a full-industry chain platform, characterized in that, include: The ChainSec Data Warehouse module is used to collect multi-source data from the entire supply chain in real time, and uses blockchain and lightweight encryption to build a distributed supply chain data warehouse. The supplier recommendation module is used to perform feature mapping and weight calculation based on dual core data, and generate a multi-dimensional matching score table to recommend suppliers. The bid evaluation report module is used to receive encrypted bid documents, decrypt and store them, calculate the comprehensive score based on quantitative indicators and expert qualitative indicators, and generate a structured bid evaluation report. The demand forecasting module is used to call multi-dimensional correlated data to build a hierarchical forecasting model and output three-dimensional forecasting results; The order generation module combines the results of three-dimensional demand forecasting, calculates the supply-demand gap using a supply-demand elasticity coefficient coupling algorithm, automatically generates orders, and synchronizes the order status to all terminals in real time. The formula for calculating the supply-demand gap is as follows: , This indicates a real-time feedback correction factor; This represents a dynamic coefficient, serving as a coupling bridge between demand elasticity and supply elasticity. When the market is dominated by demand fluctuations, its value approaches 1; when supply constraints are more significant, its value approaches 0, achieving scenario-based adaptation of supply and demand elasticity. This represents the three-dimensional demand forecast value at time t, which is the total future demand predicted by multiple factors. This represents the market volatility coefficient at time t, reflecting the intensity of market volatility on the demand side. Indicates the current point in time in the calculation; The policy adjustment coefficient at time t represents the degree of impact of changes in demand-side policies on demand. It is the first derivative of the demand forecast with respect to time, reflecting the instantaneous rate of change of demand over time; Indicates the corresponding Three-dimensional demand forecast values ​​at any given time; It is a sine function of the market volatility coefficient, simulating the periodic characteristics of market fluctuations; This represents the overall supply capacity value at time t; This represents the production elasticity coefficient at time t; Indicates the time response threshold; It is the time response threshold at time s, reflecting the speed at which the supply side responds to changes in demand; The logistics planning module is used to build a digital twin model of the supply chain, retrieve real-time logistics data, and dynamically plan the optimal transportation route through the intelligent traffic algorithm. The risk monitoring module is used to monitor risk factors in real time. When a risk signal is detected, it automatically sends early warning information, generates response strategies, integrates full-chain operational data, and displays it visually. Specifically, the use of blockchain and lightweight encryption to build a distributed supply chain data warehouse involves encrypting the processed data blocks using a lightweight encryption algorithm, and uploading the hash value and the encrypted data to the blockchain network to build the supply chain data warehouse. After receiving and processing the standardized data block, a dynamic key derivation mechanism is introduced on the basis of the national cryptographic SM4 algorithm. First, the core features in the data block tag are extracted, and a special key bound to the data features is generated through a preset key derivation function. The key is used to perform SM4 block encryption. At the same time, a dynamic factor based on timestamp is embedded in the encryption process, so that the encryption results of the same data block are different at different times. After encryption, a double hash value is calculated on the ciphertext. The main hash is based on the entire ciphertext, and the secondary hash is based on the segmented ciphertext. The double hash, key derivation parameters, and dynamic factor metadata are packaged into a blockchain transaction and written into the ledger after consensus among the consortium blockchain nodes. The encrypted data blocks are distributed and stored in an off-chain sharded cluster according to feature tags. Specifically, based on a standardized feature vector set, a dynamic hierarchical analysis model is constructed based on business scenarios, dividing the feature dimensions into target, criterion, and solution layers. The scenario feature extraction module parses current business attributes to generate a scenario factor matrix, dynamically corrects the criterion layer judgment matrix, and adjusts element values ​​by combining historical case feature influence statistical learning. The weights of each dimension are calculated using the eigenvalue method and pass a consistency check. Subsequently, a multi-dimensional matching degree calculation formula is introduced: , and These are the start time and the end time, respectively; For the index of the feature dimension, This represents the total number of feature dimensions. Indicates the first Dynamic weighting function for each feature dimension For scene parameters and It also satisfies that the sum of the weights of all dimensions is 1, and its value is dynamically adjusted according to changes in business scenarios to reflect the importance of each feature dimension in different scenarios. Indicates that the supplier in the Temporal vectors in each feature dimension As a time variable, it reflects the dynamic changes of this feature dimension over time; Indicates the first The historical influence coefficients of each feature dimension are updated in real time using a Bayesian estimation algorithm, reflecting the weight of the historical performance of that feature dimension on the current matching degree. Represents the exponentially decaying term, where yes The first derivative with respect to time, i.e., the characteristic rate of change; This represents the covariance matrix between feature dimensions, reflecting the correlation between different feature dimensions. Its determinant value is used to quantify the impact of the degree of coupling between feature dimensions on the overall matching degree. This represents the Laplacian operator, which enhances the impact of variations in the spatial distribution of features on the matching degree. This represents the feature matching deviation matrix, where the elements reflect the degree of deviation between the actual and ideal values ​​of each feature dimension. This indicates that the factor vector is adjusted in real time. The elements in the vector are dynamically generated based on real-time business data and are used to correct the impact of feature matching deviation on the final result. This represents the Hadamard product operator, used to implement element-level interactions such as feature matching bias and real-time adjustment factors. Denotes the shock response function vector, where The lag time reflects the degree of impact of external shocks on feature matching under different lag times; This represents the real-time perturbation vector. This represents the current moment corresponding to the lag time, and the elements in the vector reflect the external disturbances that occur in real time. Perform a weighted operation on the feature vectors of each supplier and their corresponding weights to generate multi-dimensional matching sub-scores and comprehensive scores. Sort the comprehensive scores and associate them with feature matching details to generate a supplier recommendation list.

9. A computer-readable storage medium, characterized in that, It includes one or more program instructions, which are executed by a processor as described in any one of claims 1-7, to provide a supply chain collaborative management method for a full-industry linkage platform.

Images