Full-link supply chain collaborative management and control landing method and system

Abstract

The application discloses a whole-link supply chain collaborative management and control landing method and system, relates to the technical field of supply chain collaborative management, and comprises the following steps: acquiring operation data of each business node of a supply chain, generating a supply chain state matrix representing the operation state of each business node based on the operation data, calculating the collaborative correlation degree between each business node based on the supply chain state matrix, and generating a collaborative entropy value representing the overall collaborative state of the supply chain; based on the collaborative entropy value, risk assessment is performed by fusing demand fluctuation information and logistics fluctuation information to generate a supply chain dynamic risk value; based on the supply chain state matrix, the collaborative entropy value and the dynamic risk value, a multi-Agent collaborative decision environment is constructed, collaborative game operation is performed by multiple decision Agents, and a candidate collaborative strategy set is generated. Through the construction of the supply chain state matrix, the calculation of the collaborative entropy, the combination of dynamic risk assessment and multi-Agent collaborative decision optimization, the real-time performance and the intelligent level of the overall collaborative management and control of the supply chain are improved.

Description

Technical Field

[0001] This invention relates to the field of supply chain collaborative management technology, and in particular to the implementation methods and systems for full-chain supply chain collaborative control. Background Technology

[0002] In the field of modern supply chain management, with the deepening of globalization and the ongoing digital transformation, the supply chain has gradually evolved from a traditional linear chain structure into a complex network system involving multiple stakeholders, multiple levels of coupling, and multiple interconnected links. To improve supply chain operational efficiency and resource allocation capabilities, enterprises typically use information platforms such as Enterprise Resource Planning (ERP), Supply Chain Management (SRM), Manufacturing Execution Systems (MES), Warehouse Management Systems (WMS), and Transportation Management Systems (TMS) to digitally manage business processes such as procurement, production, warehousing, logistics, and sales. Building on this foundation, some studies further introduce technologies such as big data analytics, intelligent forecasting, risk warning, and optimized scheduling. By collecting operational data from each node of the supply chain, they monitor and analyze inventory status, order fulfillment, logistics and transportation status, and changes in market demand to achieve dynamic perception and collaborative management of the supply chain's operational status.

[0003] However, in existing supply chain collaborative management technologies, because supply chain networks involve multiple business entities and various resource elements, the operational status of each node usually exhibits complex relationships and dynamic coupling characteristics. Most existing technologies focus on analyzing and optimizing the status of individual nodes or local business processes, and their ability to quantitatively represent the overall collaborative status of the supply chain is relatively limited. When supply chain operations experience demand fluctuations, logistics delays, inventory anomalies, or resource imbalances, existing technologies often struggle to accurately reflect the degree of collaborative changes between nodes from a full-link collaborative perspective, thus affecting the accuracy of subsequent risk assessment results and the generation of collaborative decision-making strategies. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for implementing collaborative management and control across the entire supply chain, solving the problems of difficulty in quantifying the collaborative status of the entire supply chain, difficulty in accurately assessing dynamic risks, and difficulty in optimizing collaborative strategies in real time.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, the present invention provides a method for implementing collaborative management and control across the entire supply chain, comprising, Obtain operational data from each business node in the supply chain, generate a supply chain state matrix representing the operational status of each business node based on the operational data, calculate the degree of collaborative association between each business node based on the supply chain state matrix, and generate a collaborative entropy value representing the overall collaborative status of the supply chain. Based on the collaborative entropy value, risk assessment is carried out by integrating demand fluctuation information and logistics fluctuation information to generate a dynamic risk value for the supply chain. Based on the supply chain state matrix, collaborative entropy value and dynamic risk value, a multi-agent collaborative decision-making environment is constructed. Multiple decision agents perform collaborative game calculations to generate a set of candidate collaborative strategies. The candidate collaborative strategy set is input into the pre-trained reinforcement learning decision model for strategy evaluation and iterative optimization to generate the target collaborative control strategy. Based on the target collaborative control strategy, the supply chain collaborative control operation is executed, and the operation data of each business node in the supply chain during the control process is collected and uploaded to the database for storage.

[0007] As a preferred embodiment of the end-to-end supply chain collaborative management and control method of the present invention, the step of acquiring the operational data of each business node in the supply chain and generating a supply chain state matrix representing the operational status of each business node based on the operational data includes the following steps: Identify nodes in the supply chain, including suppliers, procurement, production, warehousing, logistics, sales, and customers, and determine the business category, location, and relationships of each node. Real-time collection of operational data from each node, including inventory levels, order completion status, capacity utilization, logistics and transportation status, and delivery timeliness, and preprocessing of the collected data; Based on the preprocessed node operation data, a vector representing the current operating status of each node is generated, and a supply chain status matrix is ​​constructed according to the business relationships between nodes. Each row in the supply chain status matrix corresponds to a single node, and each column corresponds to an operational status indicator.

[0008] As a preferred embodiment of the end-to-end supply chain collaborative management and control method of the present invention, the step of calculating the degree of collaborative association between each business node based on the supply chain state matrix and generating a collaborative entropy value representing the overall collaborative state of the supply chain includes the following steps: Based on the supply chain status matrix, analyze the business relationships and data dependencies between nodes, and establish a collaborative relationship matrix between nodes; For each node in the inter-node collaboration matrix, calculate its collaboration probability with other nodes; The collaboration probabilities of each node are normalized, and based on the normalized collaboration probabilities, the collaboration entropy value of the entire supply chain is calculated according to the information entropy formula.

[0009] As a preferred embodiment of the end-to-end supply chain collaborative management and control method described in this invention, the step of performing risk assessment based on collaborative entropy value, integrating demand fluctuation information and logistics fluctuation information, and generating a dynamic supply chain risk value includes the following steps: Obtain historical demand data from each node of the supply chain and calculate demand fluctuation indicators based on the demand data; Acquire real-time logistics data from each node of the supply chain, including transportation timeliness, transportation delays, and logistics disruption events, and calculate logistics fluctuation indicators; The collaborative entropy value is integrated with demand fluctuation indicators and logistics fluctuation indicators, and linearly weighted according to preset weights to generate a dynamic risk value for the supply chain.

[0010] As a preferred embodiment of the end-to-end supply chain collaborative management and control method described in this invention, the step of constructing a multi-agent collaborative decision-making environment based on the supply chain state matrix, collaborative entropy value, and dynamic risk value, and generating a candidate collaborative strategy set through collaborative game calculations performed by multiple decision agents, includes the following steps: The supply chain state matrix, collaborative entropy value, and dynamic risk value are used as environmental states to initialize multiple decision agents, each corresponding to a different node in the supply chain. Define a set of actionable actions for each decision agent, including adjusting purchase volume, modifying production schedule, inventory allocation, optimizing logistics routes, and order fulfillment strategies; For each action in the set of executable actions, quantify the impact of each action on the overall supply chain objectives, and construct a collaborative benefit function among agents; Each decision agent performs game theory based on the environmental state and collaborative reward function. Through iterative calculation, it determines the action selection of each node that meets the threshold condition, and combines the actions of each agent to form a candidate collaborative strategy set.

[0011] As a preferred embodiment of the end-to-end supply chain collaborative management and control method of the present invention, the step of inputting the candidate collaborative strategy set into a pre-trained reinforcement learning decision model for strategy evaluation and iterative optimization to generate a target collaborative management and control strategy includes the following steps: The candidate collaborative strategy set is used as the action space input to the pre-trained reinforcement learning decision model, and the corresponding supply chain state matrix, collaborative entropy value and dynamic risk value are used as the state space input model. Based on the pre-set reward function in the reinforcement learning model, each candidate strategy is evaluated, and its impact on the overall supply chain objective is calculated. Gradient descent is used to update the parameters of the reinforcement learning decision model, enabling the model to adaptively optimize candidate policy combinations and identify the optimal or near-optimal collaborative policy in the entire process. During the strategy iteration and optimization process, the results of the reinforcement learning model are mapped to the candidate strategy set, and the strategies that meet the preset performance indicators are selected as the target collaborative management strategy.

[0012] As a preferred embodiment of the end-to-end supply chain collaborative management and control method described in this invention, the step of executing supply chain collaborative management and control operations based on the target collaborative management and control strategy, and collecting operational data from each business node in the supply chain during the management and control process, and uploading it to the database for storage, includes the following steps: The target collaborative management and control strategy is parsed into specific execution instructions for each node in the supply chain, and the parsed execution instructions are sent to each node in the supply chain to perform the corresponding operations. During the execution process, operational data from each supply chain node is collected in real time and uploaded to the database for storage.

[0013] Secondly, this invention provides a system for implementing collaborative management and control across the entire supply chain, including: The data acquisition module collects real-time operational data from each business node in the supply chain. The state matrix construction module organizes the operational data of each node into a structured supply chain state matrix; The collaborative entropy calculation module evaluates the collaborative relationship between nodes based on the state matrix and generates a collaborative entropy value that represents the overall collaborative state. The risk quantification module integrates collaborative entropy, demand fluctuation information, and logistics fluctuation information to generate dynamic risk values ​​for the supply chain. The multi-Agent decision-making module constructs a decision-making environment, enabling multiple agents to perform collaborative game based on state matrices, collaborative entropy, and dynamic risk values, and generate a set of candidate strategies. The reinforcement learning optimization module evaluates and iteratively optimizes the candidate policy set to generate a collaborative control policy for the target. The execution feedback module performs supply chain operations according to the target collaboration strategy and collects node operation data and uploads it to the database.

[0014] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the end-to-end supply chain collaborative management and control implementation method as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the end-to-end supply chain collaborative management and control implementation method as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By identifying nodes such as suppliers, procurement, production, warehousing, logistics, sales, and customers, and constructing a supply chain state matrix, dispersed node information can be accurately perceived at the overall level, improving the completeness and accuracy of global supply chain state monitoring; by analyzing the business dependencies and information flow relationships between nodes in the supply chain state matrix, establishing a node collaboration relationship matrix, and calculating supply chain collaboration entropy based on collaboration probability, a quantitative characterization and measurable analysis of the collaboration level of all nodes in the entire chain is achieved, enabling the identification of potential collaboration bottlenecks and imbalances between nodes, improving the visualization and evaluation capabilities of the overall supply chain collaboration status; by integrating collaboration entropy values, historical and predicted demand data, and real-time logistics fluctuation information, the dynamic risk value of the supply chain is calculated, enabling the monitoring of potential risks during supply chain operation. Real-time quantitative risk assessment can identify risk nodes in advance when demand fluctuates, logistics delays occur, or inventory anomalies occur, enhancing the supply chain's risk perception and dynamic response capabilities. By constructing a multi-agent decision-making environment, each agent corresponds to a different node in the supply chain and performs collaborative game calculations based on the state matrix, collaborative entropy, and dynamic risk value, realizing cross-node strategy collaborative generation. This can balance resource allocation, scheduling plans, and risk response across the entire chain, improving the overall collaborative efficiency and multi-objective optimization capabilities of the supply chain. By inputting the candidate collaborative strategy set into a pre-trained reinforcement learning model for reward function evaluation and iterative optimization, intelligent screening and dynamic optimization of collaborative strategies across the entire chain are achieved. This enables adaptive responses to complex scenarios and uncertain events, improving the intelligence level and decision-making accuracy of collaborative strategies. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the implementation of collaborative management and control methods across the entire supply chain.

[0019] Figure 2 A schematic diagram of a collaborative management and control system for the entire supply chain.

[0020] Figure 3 A flowchart for generating a set of candidate collaborative strategies.

[0021] Figure 4 A flowchart for generating the target collaborative management and control strategy execution control. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for implementing collaborative management and control of the entire supply chain, including the following steps: The system acquires operational data from each business node in the supply chain, generates a supply chain state matrix representing the operational status of each business node based on the operational data, calculates the degree of collaborative relationship between each business node based on the supply chain state matrix, and generates a collaborative entropy value representing the overall collaborative status of the supply chain.

[0026] Specifically, the nodes in the supply chain, including suppliers, procurement, production, warehousing, logistics, sales, and customers, are identified, and the business category, location, and relationship of each node are determined. Real-time collection of operational data from each node, including inventory levels, order completion status, capacity utilization, logistics and transportation status, and delivery timeliness, and preprocessing of the collected data; Based on the preprocessed node operation data, a vector representing the current operating status of each node is generated, and a supply chain status matrix is ​​constructed according to the business relationships between nodes. Each row in the supply chain status matrix corresponds to a single node, and each column corresponds to an operational status indicator.

[0027] By identifying nodes among the various business entities in the supply chain and collecting and preprocessing operational data based on node business categories, spatial locations, and relationships, the entire supply chain information has been transformed from scattered records to a unified, structured expression. This enables the correlation and analysis of inventory, orders, capacity, logistics, and delivery status at different stages within the same data framework. Furthermore, the construction of node status vectors and a supply chain status matrix provides a standardized and computable data foundation for subsequent collaborative relationship calculations, risk identification, and strategy optimization, thereby enhancing the completeness, relevance, and analyzability of supply chain operational status perception.

[0028] Furthermore, based on the supply chain status matrix, the business relationships and data dependencies between nodes are analyzed, and a collaborative relationship matrix between nodes is established. For each node in the inter-node collaboration matrix, calculate its collaboration probability with other nodes, using the following formula: ; in, Index for business nodes, This refers to another business node that has a collaborative relationship with the business node. The total number of business nodes. For business nodes With business nodes The degree of coordination between them For the first The probability of collaboration among individual business nodes.

[0029] The collaboration probabilities of each node are normalized, and based on the normalized collaboration probabilities, the collaboration entropy value of the entire supply chain is calculated according to the information entropy formula, which is: ; in, The collaborative entropy value of the entire supply chain. For the normalized first The probability of collaboration among individual business nodes.

[0030] By analyzing the business relationships and data dependencies between nodes based on the supply chain state matrix and establishing a node collaboration matrix, a quantitative representation of the node collaboration level is achieved. Subsequently, by calculating the node collaboration probability and performing normalization processing, the overall supply chain collaboration entropy value is generated based on the information entropy formula, which can accurately reflect the collaboration status of nodes across the entire chain, providing a scientific basis for supply chain collaboration optimization and risk assessment, and improving the accuracy and quantifiability of overall supply chain collaboration perception.

[0031] Based on the collaborative entropy value, risk assessment is carried out by integrating demand fluctuation information and logistics fluctuation information to generate a dynamic risk value for the supply chain. Based on the supply chain state matrix, collaborative entropy value, and dynamic risk value, a multi-agent collaborative decision-making environment is constructed. Through multiple decision agents performing collaborative game calculations, a set of candidate collaborative strategies is generated.

[0032] Specifically, historical demand data from each node in the supply chain is obtained, and a demand fluctuation index is calculated based on this data. The formula is as follows: ; in, For the first Demand fluctuation indicators for each business node For the first The standard deviation of historical demand data for each business node For the first The average of historical demand data for each business node.

[0033] Acquire real-time logistics data from each node of the supply chain, including transportation timeliness, transportation delays, and logistics disruption events, and calculate the logistics fluctuation index using the following formula: ; in, For the first Logistics fluctuation indicators for each business node For the first The standard deviation of transportation timeliness at each business node For the first The average delivery time of each business node.

[0034] The collaborative entropy value is integrated with demand fluctuation indicators and logistics fluctuation indicators, and then linearly weighted according to preset weights to generate a dynamic supply chain risk value. The formula is as follows: ; in, For the first Dynamic risk value of each business node For the first The collaborative entropy value of each business node. , , These are the weighting coefficients for the set collaborative entropy value, demand fluctuation index, and logistics fluctuation index, respectively.

[0035] By acquiring historical demand data, forecasted demand data, and real-time logistics operation data from each node of the supply chain, and constructing demand fluctuation indicators and logistics fluctuation indicators respectively, a dynamic representation of market demand change trends and logistics operation stability is achieved. This enables the comprehensive identification of potential sources of supply chain risks from both the demand and logistics sides. Furthermore, by integrating the collaborative entropy value with the demand fluctuation indicators and logistics fluctuation indicators, a correlation analysis between the supply chain collaborative status and external disturbance factors is realized. This provides a more comprehensive reflection of the risk level existing in the supply chain operation process, offering a reliable basis for subsequent collaborative decision-making and resource scheduling, and improving the comprehensiveness, real-time nature, and accuracy of supply chain risk assessment.

[0036] Furthermore, the supply chain state matrix, collaborative entropy value, and dynamic risk value are used as environmental states to initialize multiple decision agents, each corresponding to a different node in the supply chain; Define a set of actionable actions for each decision agent, including adjusting purchase volume, modifying production schedule, inventory allocation, optimizing logistics routes, and order fulfillment strategies; For each action in the set of executable actions, quantify the impact of each action on the overall supply chain objectives, and construct a collaborative benefit function among agents, the formula of which is: ; in, For decision-making agents index, Index for executable actions, This represents the total number of executable actions. For the first The collaborative benefit value of each decision agent. For the first obtained based on environmental state The agent executes the first decision. The impact of each action on the overall supply chain objectives. The selection state for the action refers to the first... The agent executes the first decision. One executable action; Each decision agent performs game theory based on the environmental state and collaborative reward function. Through iterative calculation, it determines the action selection of each node that meets the threshold condition, and combines the actions of each agent to form a candidate collaborative strategy set.

[0037] By constructing a multi-agent collaborative decision-making environment from the supply chain state matrix, collaborative entropy value, and dynamic risk value, and defining executable actions and collaborative benefit functions for each agent, it is possible to enable each node to make autonomous strategy selection under the overall goal of the supply chain. Through game theory, a set of candidate collaborative strategies is generated, which enables the resource scheduling, task allocation, and risk response of each node in the supply chain to be dynamically optimized under a unified collaborative framework, thereby improving the intelligence, collaboration, and global operational efficiency of the overall supply chain decision-making.

[0038] The candidate collaborative strategy set is input into the pre-trained reinforcement learning decision model for strategy evaluation and iterative optimization to generate the target collaborative control strategy. Based on the target collaborative control strategy, the supply chain collaborative control operation is executed, and the operation data of each business node in the supply chain during the control process is collected and uploaded to the database for storage.

[0039] Specifically, the candidate collaborative strategy set is used as the action space input to the pre-trained reinforcement learning decision model, and the corresponding supply chain state matrix, collaborative entropy value and dynamic risk value are used as the state space input model. Based on the pre-set reward function in the reinforcement learning model, each candidate strategy is evaluated, and its impact on the overall supply chain objective is calculated. Gradient descent is used to update the parameters of the reinforcement learning decision model, enabling the model to adaptively optimize candidate policy combinations and identify the optimal or near-optimal collaborative policy in the entire process. During the strategy iteration and optimization process, the results of the reinforcement learning model are mapped to the candidate strategy set, and the strategies that meet the preset performance indicators are selected as the target collaborative management strategy.

[0040] By using the candidate collaborative strategy set as the action space and the supply chain state matrix, collaborative entropy value, and dynamic risk value as the state space input to the reinforcement learning decision model, dynamic correlation analysis between supply chain operation state and decision strategy is achieved. This enables candidate strategies to be evaluated in a targeted manner under the constraints of real supply chain operation environment characteristics. Furthermore, the impact of each candidate strategy on the overall supply chain goal is quantified through a reward function, and the strategy selection process is continuously optimized using a model iteration update mechanism. This achieves the goal of automatically mining near-optimal collaborative solutions from a large number of candidate strategies, improving the adaptability and optimization capability of collaborative decision-making in complex scenarios. At the same time, by mapping and filtering the model identification results with the candidate strategy set, the decision results are accurately matched with the supply chain operation requirements, providing a more reasonable execution basis for subsequent collaborative management and control operations, and improving the intelligence level, decision accuracy, and overall management and control efficiency of supply chain collaborative strategy generation.

[0041] Furthermore, the target collaborative control strategy is parsed into specific execution instructions for each supply chain node, and the parsed execution instructions are sent to each supply chain node to perform the corresponding operations; During the execution process, operational data from each supply chain node is collected in real time and uploaded to the database for storage.

[0042] By parsing the target collaborative management and control strategy into specific execution instructions for each supply chain node and issuing them for execution, while simultaneously collecting operational data from each node in real time and uploading it to the database, dynamic monitoring and data closure of the strategy execution process are achieved. This enables the execution effect of each node's operation to be perceived and recorded in real time, providing reliable feedback for subsequent strategy optimization and model updates, and improving the traceability of supply chain collaborative management and control operations and the system's continuous optimization capabilities.

[0043] This embodiment also provides a system for implementing collaborative management and control of the entire supply chain, including: The data acquisition module collects real-time operational data from each business node in the supply chain. The state matrix construction module organizes the operational data of each node into a structured supply chain state matrix; The collaborative entropy calculation module evaluates the collaborative relationship between nodes based on the state matrix and generates a collaborative entropy value that represents the overall collaborative state. The risk quantification module integrates collaborative entropy, demand fluctuation information, and logistics fluctuation information to generate dynamic risk values ​​for the supply chain. The multi-Agent decision-making module constructs a decision-making environment, enabling multiple agents to perform collaborative game based on state matrices, collaborative entropy, and dynamic risk values, and generate a set of candidate strategies. The reinforcement learning optimization module evaluates and iteratively optimizes the candidate policy set to generate a collaborative control policy for the target. The execution feedback module performs supply chain operations according to the target collaboration strategy and collects node operation data and uploads it to the database.

[0044] This embodiment also provides a computer device applicable to the implementation of the end-to-end supply chain collaborative management and control method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the end-to-end supply chain collaborative management and control method proposed in the above embodiment.

[0045] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0046] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the method for achieving collaborative management and control of the entire supply chain as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0047] In summary, this invention improves the completeness and accuracy of global supply chain status monitoring by: identifying nodes such as suppliers, procurement, production, warehousing, logistics, sales, and customers, and constructing a supply chain status matrix; enabling accurate perception of dispersed node information at the overall level; analyzing the business dependencies and information flow relationships between nodes in the supply chain status matrix, establishing a node collaboration relationship matrix, and calculating supply chain collaboration entropy based on collaboration probability; achieving quantitative characterization and measurable analysis of the collaboration level of all nodes in the entire chain, identifying potential collaboration bottlenecks and imbalances between nodes, and improving the visualization and evaluation capabilities of the overall supply chain collaboration status; and calculating the dynamic risk value of the supply chain by integrating collaboration entropy values, historical and predicted demand data, and real-time logistics fluctuation information, thereby enabling the monitoring of potential risks during supply chain operation. Real-time quantitative assessment of risks can identify risk nodes in advance when demand fluctuates, logistics delays occur, or inventory anomalies occur, enhancing the supply chain's risk perception and dynamic response capabilities. By constructing a multi-agent decision-making environment, each agent corresponds to a different node in the supply chain and performs collaborative game calculations based on the state matrix, collaborative entropy, and dynamic risk value, realizing cross-node strategy collaborative generation. This can balance resource allocation, scheduling plans, and risk response across the entire chain, improving the overall collaborative efficiency and multi-objective optimization capabilities of the supply chain. By inputting the candidate collaborative strategy set into a pre-trained reinforcement learning model for reward function evaluation and iterative optimization, intelligent screening and dynamic optimization of collaborative strategies across the entire chain are achieved. This enables adaptive responses to complex scenarios and uncertain events, improving the intelligence level and decision-making accuracy of collaborative strategies.

[0048] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for implementing collaborative management and control across the entire supply chain, characterized in that: include, Obtain operational data from each business node in the supply chain, generate a supply chain state matrix representing the operational status of each business node based on the operational data, calculate the degree of collaborative association between each business node based on the supply chain state matrix, and generate a collaborative entropy value representing the overall collaborative status of the supply chain. Based on the collaborative entropy value, risk assessment is carried out by integrating demand fluctuation information and logistics fluctuation information to generate a dynamic risk value for the supply chain. Based on the supply chain state matrix, collaborative entropy value and dynamic risk value, a multi-agent collaborative decision-making environment is constructed. Multiple decision agents perform collaborative game calculations to generate a set of candidate collaborative strategies. The candidate collaborative strategy set is input into the pre-trained reinforcement learning decision model for strategy evaluation and iterative optimization to generate the target collaborative control strategy. Based on the target collaborative control strategy, the supply chain collaborative control operation is executed, and the operation data of each business node in the supply chain during the control process is collected and uploaded to the database for storage.

2. The method for implementing end-to-end supply chain collaborative management as described in claim 1, characterized in that: The process of acquiring operational data from each business node in the supply chain and generating a supply chain state matrix representing the operational status of each business node based on the operational data includes the following steps: Identify nodes in the supply chain, including suppliers, procurement, production, warehousing, logistics, sales, and customers, and determine the business category, location, and relationships of each node. Real-time collection of operational data from each node, including inventory levels, order completion status, capacity utilization, logistics and transportation status, and delivery timeliness, and preprocessing of the collected data; Based on the preprocessed node operation data, a vector representing the current operating status of each node is generated, and a supply chain status matrix is ​​constructed according to the business relationships between nodes. Each row in the supply chain status matrix corresponds to a single node, and each column corresponds to an operational status indicator.

3. The method for implementing end-to-end supply chain collaborative management as described in claim 2, characterized in that: The process of calculating the degree of collaborative correlation between business nodes based on the supply chain state matrix and generating a collaborative entropy value representing the overall collaborative state of the supply chain includes the following steps: Based on the supply chain status matrix, analyze the business relationships and data dependencies between nodes, and establish a collaborative relationship matrix between nodes; For each node in the inter-node collaboration matrix, calculate its collaboration probability with other nodes; The collaboration probabilities of each node are normalized, and based on the normalized collaboration probabilities, the collaboration entropy value of the entire supply chain is calculated according to the information entropy formula.

4. The method for implementing end-to-end supply chain collaborative management as described in claim 3, characterized in that: The process of assessing risk based on collaborative entropy values, integrating demand fluctuation information and logistics fluctuation information, and generating dynamic supply chain risk values ​​includes the following steps: Obtain historical demand data from each node of the supply chain and calculate demand fluctuation indicators based on the demand data; Acquire real-time logistics data from each node of the supply chain, including transportation timeliness, transportation delays, and logistics disruption events, and calculate logistics fluctuation indicators; The collaborative entropy value is integrated with demand fluctuation indicators and logistics fluctuation indicators, and linearly weighted according to preset weights to generate a dynamic risk value for the supply chain.

5. The method for implementing end-to-end supply chain collaborative management as described in claim 4, characterized in that: The process involves constructing a multi-agent collaborative decision-making environment based on the supply chain state matrix, collaborative entropy value, and dynamic risk value. Multiple decision agents then perform collaborative game calculations to generate a candidate collaborative strategy set. This includes the following steps: The supply chain state matrix, collaborative entropy value, and dynamic risk value are used as environmental states to initialize multiple decision agents, each corresponding to a different node in the supply chain. Define a set of actionable actions for each decision agent, including adjusting purchase volume, modifying production schedule, inventory allocation, optimizing logistics routes, and order fulfillment strategies; For each action in the set of executable actions, quantify the impact of each action on the overall supply chain objectives, and construct a collaborative benefit function among agents; Each decision agent performs game theory based on the environmental state and collaborative reward function. Through iterative calculation, it determines the action selection of each node that meets the threshold condition, and combines the actions of each agent to form a candidate collaborative strategy set.

6. The method for implementing end-to-end supply chain collaborative management as described in claim 5, characterized in that: The process of inputting the candidate collaborative strategy set into a pre-trained reinforcement learning decision model for strategy evaluation and iterative optimization to generate a target collaborative management strategy includes the following steps: The candidate collaborative strategy set is used as the action space input to the pre-trained reinforcement learning decision model, and the corresponding supply chain state matrix, collaborative entropy value and dynamic risk value are used as the state space input model. Based on the pre-set reward function in the reinforcement learning model, each candidate strategy is evaluated, and its impact on the overall supply chain objective is calculated. Gradient descent is used to update the parameters of the reinforcement learning decision model, enabling the model to adaptively optimize candidate policy combinations and identify the optimal or near-optimal collaborative policy in the entire process. During the strategy iteration and optimization process, the results of the reinforcement learning model are mapped to the candidate strategy set, and the strategies that meet the preset performance indicators are selected as the target collaborative management strategy.

7. The method for implementing end-to-end supply chain collaborative management as described in claim 6, characterized in that: The process of executing supply chain collaborative management and control operations based on the target collaborative management and control strategy, and collecting operational data from each business node in the supply chain during the management and control process, and uploading it to the database for storage, includes the following steps: The target collaborative management and control strategy is parsed into specific execution instructions for each node in the supply chain, and the parsed execution instructions are sent to each node in the supply chain to perform the corresponding operations. During the execution process, operational data from each supply chain node is collected in real time and uploaded to the database for storage.

8. A system for implementing collaborative management and control of the entire supply chain, based on the method for implementing collaborative management and control of the entire supply chain as described in any one of claims 1 to 7, characterized in that: include, The data acquisition module collects real-time operational data from each business node in the supply chain. The state matrix construction module organizes the operational data of each node into a structured supply chain state matrix; The collaborative entropy calculation module evaluates the collaborative relationship between nodes based on the state matrix and generates a collaborative entropy value that represents the overall collaborative state. The risk quantification module integrates collaborative entropy, demand fluctuation information, and logistics fluctuation information to generate dynamic risk values ​​for the supply chain. The multi-Agent decision-making module constructs a decision-making environment, enabling multiple agents to perform collaborative game based on state matrices, collaborative entropy, and dynamic risk values, and generate a set of candidate strategies. The reinforcement learning optimization module evaluates and iteratively optimizes the candidate policy set to generate a collaborative control policy for the target. The execution feedback module performs supply chain operations according to the target collaboration strategy and collects node operation data and uploads it to the database.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the end-to-end supply chain collaborative management and control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the end-to-end supply chain collaborative management and control method as described in any one of claims 1 to 7.

Images