A supply chain evolution game decision optimization method and system fusing multi-source data

Abstract

The present application belongs to the technical field of supply chain decision optimization, and particularly relates to a supply chain evolution game decision optimization method and system fusing multi-source data. The present application firstly calculates strategy distribution entropy based on the strategy selection proportion of each subject, then calculates external disturbance factors based on external environment indexes in the global trusted state data, secondly calculates comprehensive game state entropy based on the calculation results of strategy distribution entropy and the calculation results of external disturbance factors, and finally sets a game imbalance early warning threshold to pre-judge potential risks based on the calculation results of comprehensive game state entropy, and automatically generates a regulation and control strategy to intervene in the game process when the early warning is triggered, thereby guaranteeing the stable operation of the supply chain. This progressive calculation method makes the entropy value not only reflect the change of the internal game state, but also perceive the disturbance of the external environment, thereby realizing accurate risk measurement combining internal and external factors, and being more scientific and authentic than single-dimensional entropy value calculation.

Description

Technical Field

[0001] This invention belongs to the field of supply chain decision optimization technology, specifically relating to a supply chain evolution game decision optimization method and system that integrates multi-source data. Background Technology

[0002] In the context of the current globalization and digital economy, supply chain systems face high uncertainty and dynamism. Traditional supply chain decision-making methods usually rely on single-dimensional historical data for demand forecasting, which is difficult to cope with the complex disturbances brought about by multi-source heterogeneous data. In addition, the supply chain is composed of multiple independent entities with bounded rationality. When each entity pursues the maximization of its own interests, it will generate complex game behavior. Although there are separate applications of data fusion or game theory in existing technologies, there is a lack of a decision-making mechanism that can deeply couple the real-time fusion of multi-source data with the dynamic game evolution between entities. This results in a delayed response of the decision-making system, low coordination efficiency, and an inability to effectively warn and control the imbalance of the game.

[0003] To address the aforementioned issues, this application presents a supply chain evolution game decision optimization method and system that integrates multi-source data. Summary of the Invention

[0004] To address the shortcomings of existing technologies mentioned in the background section, this application proposes a supply chain evolutionary game decision optimization method and system that integrates multi-source data. First, the invention introduces an improved replicative dynamic equation to solve the strategy evolution problem through a constructed game model, seeking an evolutionarily stable strategy. Then, iteratively corrects the model parameters based on real-time data streams. Next, it generates differentiated decision schemes based on the evolutionarily stable strategy, establishes a decision performance evaluation index system, and generates feedback data by comparing actual execution results with expected goals, driving the model to perform closed-loop parameter optimization. Finally, it calculates the comprehensive game state entropy based on the strategy selection ratios of each subject and external environmental indicators in the global credible state data. Finally, based on the calculation results of the comprehensive game state entropy, it sets a game imbalance warning threshold to predict potential risks. When a warning is triggered, it automatically generates a control strategy to intervene in the game process, ensuring the stable operation of the supply chain and solving the problems in the background section.

[0005] Firstly, to achieve the above objectives, this application provides a supply chain evolution game decision optimization method that integrates multi-source data, which includes the following specific steps: S1: Collect multi-source heterogeneous data from the supply chain, and perform cleaning, noise reduction, standardization, and trust verification on the collected multi-source heterogeneous data to generate fused global trustworthy status data. S2: Based on the global credible state data output in step S1, identify the game players and their strategy space in the supply chain, construct an improved dynamic evolutionary game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function, and reward / penalty coefficients. S3: Based on the game model constructed in step S2, an improved replicating dynamic equation is introduced to solve the strategy evolution, find the evolutionary stable strategy, and iteratively correct the model parameters by combining real-time data streams; S4: Generate differentiated decision-making schemes based on the evolutionary stable strategy obtained in step S3, establish a decision-making effect evaluation index system, and generate feedback data by comparing the actual execution effect with the expected goal to drive the model to optimize closed-loop parameters; S5: Calculate the state entropy of the comprehensive game based on the strategy selection ratio of each subject and the external environment indicators in the global credible state data; The calculation of the comprehensive game state entropy Qt based on the strategy selection ratios of each agent and external environment indicators in the global credible state data specifically includes the following steps: S51: The strategy distribution entropy is calculated based on the strategy selection ratio of each subject. The formula for calculating the strategy distribution entropy is: ; Where Ht is the strategy distribution entropy at time t, which measures the degree of dispersion of strategy choices among various entities in the supply chain. A larger value indicates a more uniform and chaotic strategy selection, while a smaller value indicates a more concentrated and rigid strategy. Let be the proportion of players in group i who choose strategy a at time t, N be the total number of players in the supply chain, and K be the total number of strategies that a single player can choose. This represents taking the logarithm of the policy proportion, used to quantify information uncertainty; S52: The external disturbance factor is calculated based on the external environment indicators in the global trusted state data. The formula for calculating the external disturbance factor is as follows: ; Where Dt is the external disturbance factor at time t, used to measure the overall volatility of the external environment indicators of the supply chain; the larger the value, the more severe the environmental disturbance. Let pj be the real-time value of the j-th external environmental indicator, zj be the historical mean of the j-th external environmental indicator, zj be the historical standard deviation of the j-th external environmental indicator, and M be the number of external indicators. S53: Calculate the state entropy of the comprehensive game based on the calculation results of the strategy distribution entropy and the calculation results of the external disturbance factor; S6: Based on the calculation results of the state entropy of the comprehensive game, a game imbalance early warning threshold is set to predict potential risks. When the early warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain.

[0006] Based on the above scheme, the preferred trusted verification process in step S1 specifically includes: for multi-source heterogeneous data from the same supply chain, calculating the historical accuracy of each data source as the confidence weight, and generating a fused trusted data value through a weighted average method.

[0007] Based on the above scheme, the adaptive initialization formula for the profit function in step S2 is preferred as follows: ; in, Let be the adaptive payoff of agents i and j at time t when they adopt the strategy combination (a, b). This is the core decision-making basis for agents choosing strategies in evolutionary game theory, and it updates dynamically with time and the environment. t is a time variable, representing different moments in the dynamic evolution of the supply chain, reflecting the real-time nature of the payoff. is the basic revenue constant for the strategy combination (a, b), which is the benchmark revenue without considering demand fluctuations, costs, and synergistic rewards and penalties. It is pre-set by the characteristics of the supply chain business. y is the demand impact coefficient, used to adjust the strength of the impact of real-time demand fluctuations on the basic revenue. The real-time cost of agent i at time t when it adopts strategy a includes dynamic costs such as production costs, inventory costs, and logistics costs that change with time and strategy. It is calculated directly from multi-source fusion data in real time. i is the game agent number representing a participant in the supply chain, such as a supplier, manufacturer, or distributor. a is the strategy number adopted by agent i, such as high inventory strategy, low inventory strategy, or collaborative supply strategy. The reward / penalty coefficient between subjects i and j at time t is dynamically calculated based on the real-time market supply and demand ratio. When supply exceeds demand, the penalty for non-cooperative strategies is lowered; when supply is less than demand, the reward for cooperative strategies is raised. Let be the base value of the synergistic benefit of the strategy combination (a, b). It represents the fixed additional benefit or loss benchmark generated by cooperation such as information sharing and joint replenishment when subjects i and j adopt strategy (a, b). j is the number of another subject playing against subject i, representing a partner or competitor in the supply chain. b is the strategy number adopted by subject j, which together with subject i's strategy a constitutes the strategy combination (a, b).

[0008] Based on the above scheme, the improved formula for the replication dynamic equation in step S3 is as follows: ; in, The proportion of the main group i that chooses strategy a. To determine the expected return of strategy a, The average expected return of the population is given by numerically solving the replication dynamic equation until convergence, thus obtaining the evolutionary stable policy, where dt is a small time increment.

[0009] Based on the above scheme, the preferred decision-making effect evaluation index system in step S4 includes on-time delivery rate, total supply chain operating cost and inventory level volatility index. By comprehensively comparing the deviation between the actual execution effect and the expected goal, the game model parameters in step S2 are triggered to optimize and adjust in a closed loop.

[0010] Based on the above scheme, the preferred formula for calculating the comprehensive game state entropy is: ; Where Qt is the comprehensive game state entropy at time t, which is used to simultaneously characterize the comprehensive risk level of internal strategy structure and external environmental disturbances. It is the core basis for risk warning. α is the adjustment coefficient, which is used to control the influence weight of external disturbances on the comprehensive entropy. The larger the value of α, the stronger the influence of external disturbance factors on the system state.

[0011] Based on the above scheme, the preferred embodiment, which uses the calculation results of the comprehensive game state entropy to set a game imbalance early warning threshold to predict potential risks, and automatically generates a control strategy to intervene in the game process when the early warning is triggered, thereby ensuring the stable operation of the supply chain, includes the following steps: S61. Set the lower threshold Qmin and the upper threshold Qmax for the state entropy of the comprehensive game. S62. Risk assessment is based on the relationship between the state entropy Qt of the comprehensive game and the threshold. This indicates that the current supply chain faces a risk of strategic rigidity. This indicates that there is a risk of strategic confusion in the current supply chain; S63. Implement corresponding dynamic adjustment strategies based on different risk types. When the risk of strategy chaos is triggered, execute the first adjustment strategy to temporarily increase the collaborative return term in the return function. When the risk of policy rigidity is triggered, a second control policy is executed, introducing a random perturbation term δ into the replication dynamic equation, resulting in an improved replication dynamic equation: ; Wherein, δ is a random perturbation term used to break policy fixation and explore new policy spaces; S64. Continuously monitor the state entropy Qt of the comprehensive game until it returns to the safe range, and then gradually withdraw the control measures to restore the normal operation of the system.

[0012] Secondly, this application provides a supply chain evolution game decision optimization system that integrates multi-source data, which specifically includes: a data acquisition and fusion module, a game model construction module, a dynamic deduction and optimization module, a decision generation closed-loop module, an entropy calculation module, a risk warning and control module, and a central control module; The data acquisition and fusion module is used to collect multi-source heterogeneous data from the supply chain, and to clean, reduce noise, standardize and verify the collected multi-source heterogeneous data to generate fused global reliable status data. The game model construction module is used to identify the game subjects and their strategy space in the supply chain based on the global credible state data output in step S1, construct an improved dynamic evolution game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function and reward / penalty coefficients. The dynamic deduction and optimization module is used to solve the strategy evolution problem based on the game model constructed in step S2. It introduces an improved replication dynamic equation to find an evolution-stable strategy and iteratively corrects the model parameters by combining real-time data streams. The decision generation closed-loop module is used to execute step S4 to complete the decision generation, evaluation and closed-loop optimization functions; The entropy calculation module is used to perform the comprehensive game state entropy calculation function described in step S5; The risk warning and control module is used to perform risk warning and dynamic control functions in step S6; The central control module is connected to the data acquisition and fusion module, the game model construction module, the dynamic inference and optimization module, the decision generation closed-loop module, the entropy calculation module, and the risk early warning and control module to coordinate the work of each module and the data interaction.

[0013] Thirdly, this application provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the aforementioned supply chain evolution game decision optimization method that integrates multi-source data by calling the computer program stored in the memory.

[0014] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned supply chain evolution game decision optimization method that integrates multi-source data.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves synchronous updates between the game model and the real-time state of the supply chain through the game model construction and parameter adaptive initialization in step S2, and the evolutionary game dynamic deduction and optimization in step S3. The real-time demand fluctuation factor, real-time cost, and dynamic reward and punishment coefficient in the payoff function can all be obtained in real time from the global reliable state data output in step S1 and adaptively adjusted, so that the game model fits the actual operating state of the current supply chain from the initial moment. At the same time, the model parameters are iteratively corrected in step S3 by combining real-time data streams, ensuring dynamic synchronization between the deduction process and changes in the external environment, overcoming the defect that traditional static game models cannot adapt to environmental changes. This invention establishes a complete closed-loop mechanism of decision-making, execution, evaluation, feedback, and optimization through decision generation, evaluation, and closed-loop optimization in step S4. By establishing a multi-dimensional decision effect evaluation index system, the actual execution effect of the decision plan can be objectively quantified. When the actual effect deviates from the expected goal, the system can automatically trigger parameter optimization and adjust the model parameters in step S2, so that subsequent decisions are more inclined to improve overall performance. This closed-loop mechanism gives the system the ability to learn and correct itself, enabling the quality of decision-making to continuously improve with the accumulation of operational experience, demonstrating significant intelligent features. This invention first calculates the strategy distribution entropy based on the strategy selection ratio of each subject, then calculates the external disturbance factor based on the external environment indicators in the global credible state data, next calculates the comprehensive game state entropy based on the calculation results of the strategy distribution entropy and the calculation results of the external disturbance factor, and finally sets a game imbalance warning threshold based on the calculation results of the comprehensive game state entropy to predict potential risks. When the warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain. This progressive calculation method enables the entropy value to reflect changes in the internal game state and perceive disturbances in the external environment, realizing accurate risk measurement by combining internal and external factors, which is more scientific and realistic than the single-dimensional entropy value calculation. This invention, through risk warning and dynamic control in step S6, establishes an upper and lower dual-threshold warning mechanism based on the comprehensive game state entropy, and designs differentiated control strategies for different types of risks. When the risk of policy confusion is triggered, the system can be guided to evolve towards a cooperative strategy by increasing the synergistic benefit item, thus promoting the system's return to stability. When the risk of strategy rigidity is triggered, the strategy lock-in state is broken by introducing random disturbance terms, which stimulates the vitality of the system. This classification-based early warning and precise policy implementation control mechanism makes risk intervention more targeted, effectively avoids the secondary problems that may be caused by one-size-fits-all control, and significantly improves the robustness and risk resistance of the supply chain. This invention constructs a complete technical closed loop through six interconnected steps: data fusion, model self-starting, dynamic inference, decision optimization, entropy calculation, and risk control. Each step has a strict logical dependency and data transmission relationship, and none can be omitted, forming an organic whole. This technical solution realizes intelligent, precise, and efficient decision-making in supply chain evolution game theory, which can significantly improve the overall operational efficiency and stability of the supply chain, reduce operating costs and risk losses, and has significant industrial application value and socio-economic benefits. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the overall process of a supply chain evolution game decision optimization method that integrates multi-source data according to the present invention; Figure 2 This is a flowchart illustrating step S5 in the supply chain evolution game decision optimization method that integrates multi-source data according to the present invention. Figure 3 This is a schematic diagram of the framework of a supply chain evolution game decision optimization system that integrates multi-source data according to the present invention. Detailed Implementation

[0017] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 To address the technical problems raised in the background art, this application provides a preferred embodiment: such as Figure 1 - Figure 3 As shown, a supply chain evolution game decision optimization method integrating multi-source data includes the following specific steps: S1: Collect multi-source heterogeneous data from the supply chain, and perform cleaning, noise reduction, standardization, and trust verification on the collected multi-source heterogeneous data to generate fused global trustworthy status data. S2: Based on the global credible state data output in step S1, identify the game players and their strategy space in the supply chain, construct an improved dynamic evolutionary game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function, and reward / penalty coefficients. S3: Based on the game model constructed in step S2, an improved replicating dynamic equation is introduced to solve the strategy evolution, find the evolutionary stable strategy, and iteratively correct the model parameters by combining real-time data streams; S4: Generate differentiated decision-making schemes based on the evolutionary stable strategy obtained in step S3, establish a decision-making effect evaluation index system, and generate feedback data by comparing the actual execution effect with the expected goal to drive the model to optimize closed-loop parameters; S5: Calculate the state entropy of the comprehensive game based on the strategy selection ratio of each subject and the external environment indicators in the global credible state data; The calculation of the comprehensive game state entropy Qt based on the strategy selection ratios of each agent and external environment indicators in the global credible state data specifically includes the following steps: S51: The strategy distribution entropy is calculated based on the strategy selection ratio of each subject. The formula for calculating the strategy distribution entropy is: ; Where Ht is the strategy distribution entropy at time t, which measures the degree of dispersion of strategy choices among various entities in the supply chain. A larger value indicates a more uniform and chaotic strategy selection, while a smaller value indicates a more concentrated and rigid strategy. Let be the proportion of players in group i who choose strategy a at time t, N be the total number of players in the supply chain, and K be the total number of strategies that a single player can choose. This represents taking the logarithm of the policy proportion, used to quantify information uncertainty; S52: The external disturbance factor is calculated based on the external environment indicators in the global trusted state data. The formula for calculating the external disturbance factor is as follows: ; Where Dt is the external disturbance factor at time t, used to measure the overall volatility of the external environment indicators of the supply chain; the larger the value, the more severe the environmental disturbance. Let pj be the real-time value of the j-th external environmental indicator, zj be the historical mean of the j-th external environmental indicator, zj be the historical standard deviation of the j-th external environmental indicator, and M be the number of external indicators. S53: Calculate the state entropy of the comprehensive game based on the calculation results of the strategy distribution entropy and the calculation results of the external disturbance factor; S6: Based on the calculation results of the state entropy of the comprehensive game, a game imbalance early warning threshold is set to predict potential risks. When the early warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain.

[0019] The advantages of this embodiment compared to the prior art are as follows: This invention constructs a complete technical closed loop through six interconnected steps, including data fusion, model self-starting, dynamic inference, decision optimization, entropy calculation, and risk control. Each step has a strict logical dependency and data transmission relationship, and none can be omitted, forming an organic whole. This technical solution realizes the intelligent, precise, and efficient decision-making of supply chain evolution game theory, which can significantly improve the overall operating efficiency and stability of the supply chain, reduce operating costs and risk losses, and has significant industrial application value and socio-economic benefits.

[0020] Furthermore: In an optional embodiment, the trusted verification process in step S1 specifically includes: for multi-source heterogeneous data from the same supply chain, calculating the historical accuracy of each data source as a confidence weight, and generating a fused trusted data value through a weighted average method.

[0021] In an optional embodiment, the adaptive initialization formula for the profit function in step S2 is: ; in, Let be the adaptive payoff of agents i and j at time t when they adopt the strategy combination (a, b). This is the core decision-making basis for agents choosing strategies in evolutionary game theory, and it updates dynamically with time and the environment. t is a time variable, representing different moments in the dynamic evolution of the supply chain, reflecting the real-time nature of the payoff. is the basic revenue constant for the strategy combination (a, b), which is the benchmark revenue without considering demand fluctuations, costs, and synergistic rewards and penalties. It is pre-set by the characteristics of the supply chain business. y is the demand impact coefficient, used to adjust the strength of the impact of real-time demand fluctuations on the basic revenue. The real-time cost of agent i at time t when it adopts strategy a includes dynamic costs such as production costs, inventory costs, and logistics costs that change with time and strategy. It is calculated directly from multi-source fusion data in real time. i is the game agent number representing a participant in the supply chain, such as a supplier, manufacturer, or distributor. a is the strategy number adopted by agent i, such as high inventory strategy, low inventory strategy, or collaborative supply strategy. The reward / penalty coefficient between subjects i and j at time t is dynamically calculated based on the real-time market supply and demand ratio. When supply exceeds demand, the penalty for non-cooperative strategies is lowered; when supply is less than demand, the reward for cooperative strategies is raised. Let be the base value of the synergistic benefit of the strategy combination (a, b). It represents the fixed additional benefit or loss benchmark generated by cooperation such as information sharing and joint replenishment when subjects i and j adopt strategy (a, b). j is the number of another subject playing against subject i, representing a partner or competitor in the supply chain. b is the strategy number adopted by subject j, which together with subject i's strategy a constitutes the strategy combination (a, b).

[0022] In an optional embodiment, the improved replication dynamic equation in step S3 is formulated as follows: ; in, The proportion of the main group i that chooses strategy a. To determine the expected return of strategy a, The average expected return of the population is given by numerically solving the replication dynamic equation until convergence, thus obtaining the evolutionary stable policy, where dt is a small time increment.

[0023] It should be noted that: To determine the expected return of strategy a, To obtain the average expected return of the population, the Runge-Kutta method is used to numerically solve the above differential equations and iteratively calculate the evolutionary stable strategy. Furthermore, during the deduction process, the model parameters are updated whenever new real-time data is received, so as to achieve synchronous iterative correction of the model parameters and real-time data.

[0024] In an optional embodiment, the decision-making effect evaluation index system in step S4 includes on-time delivery rate, total supply chain operating cost and inventory level volatility index. By comprehensively comparing the deviation between the actual execution effect and the expected goal, the game model parameters in step S2 are triggered to optimize and adjust in a closed loop.

[0025] The advantages of this embodiment compared to the prior art are as follows: Through decision generation, evaluation, and closed-loop optimization in step S4, a complete closed-loop mechanism of decision-making, execution, evaluation, feedback, and optimization is established. By establishing a multi-dimensional decision effect evaluation index system, the actual execution effect of the decision plan can be objectively quantified. When the actual effect deviates from the expected goal, the system can automatically trigger parameter optimization and adjust the model parameters in step S2, so that subsequent decisions are more inclined to improve overall performance. This closed-loop mechanism gives the system the ability to learn and correct itself, enabling the quality of decision-making to continuously improve with the accumulation of operational experience, demonstrating significant intelligent characteristics.

[0026] Furthermore: In an optional embodiment, the formula for calculating the state entropy of the comprehensive game is: ; Where Qt is the comprehensive game state entropy at time t, which is used to simultaneously characterize the comprehensive risk level of internal strategy structure and external environmental disturbances. It is the core basis for risk warning. α is the adjustment coefficient, which is used to control the influence weight of external disturbances on the comprehensive entropy. The larger the value of α, the stronger the influence of external disturbance factors on the system state.

[0027] In an optional embodiment, based on the calculation result of the comprehensive game state entropy, a game imbalance early warning threshold is set to predict potential risks. When the early warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain, including the following steps: S61. Set the lower threshold Qmin and the upper threshold Qmax for the state entropy of the comprehensive game. S62. Risk assessment is based on the relationship between the state entropy Qt of the comprehensive game and the threshold. This indicates that the current supply chain faces a risk of strategic rigidity. This indicates that there is a risk of strategic confusion in the current supply chain; S63. Implement corresponding dynamic adjustment strategies based on different risk types. When the risk of strategy chaos is triggered, execute the first adjustment strategy to temporarily increase the collaborative return term in the return function. When the risk of policy rigidity is triggered, a second control policy is executed, introducing a random perturbation term δ into the replication dynamic equation, resulting in an improved replication dynamic equation: ; Wherein, δ is a random perturbation term used to break policy fixation and explore new policy spaces; S64. Continuously monitor the state entropy Qt of the comprehensive game until it returns to the safe range, and then gradually withdraw the control measures to restore the normal operation of the system.

[0028] The advantages of this embodiment compared to the prior art are: a dual-threshold early warning mechanism is set based on the comprehensive game state entropy, and differentiated control strategies are designed for different types of risks. When the risk of policy confusion is triggered, the system can be guided to evolve towards a cooperative strategy by increasing the synergistic benefit item, thus promoting the system's return to stability. When the risk of strategy rigidity is triggered, the system can break the strategy lock-in state by introducing random disturbance terms, thereby stimulating the system's vitality. This classification-based early warning and precise policy implementation mechanism makes risk intervention more targeted, effectively avoids secondary problems that may be caused by one-size-fits-all regulation, and significantly improves the robustness and risk resistance of the supply chain.

[0029] Example 2 Based on the same inventive concept as in Embodiment 1, such as Figure 1 As shown, this embodiment provides a supply chain evolution game decision optimization system that integrates multi-source data, which specifically includes: a data acquisition and fusion module, a game model construction module, a dynamic inference and optimization module, a decision generation closed-loop module, an entropy calculation module, a risk warning and control module, and a central control module; The data acquisition and fusion module is used to collect multi-source heterogeneous data from the supply chain, and to clean, reduce noise, standardize and verify the collected multi-source heterogeneous data to generate fused global reliable status data. The game model construction module is used to identify the game subjects and their strategy space in the supply chain based on the global credible state data output in step S1, construct an improved dynamic evolution game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function and reward / penalty coefficients. The dynamic deduction and optimization module is used to solve the strategy evolution problem based on the game model constructed in step S2. It introduces an improved replication dynamic equation to find an evolution-stable strategy and iteratively corrects the model parameters by combining real-time data streams. The decision generation closed-loop module is used to execute step S4 to complete the decision generation, evaluation and closed-loop optimization functions; The entropy calculation module is used to perform the comprehensive game state entropy calculation function described in step S5; The risk warning and control module is used to perform risk warning and dynamic control functions in step S6; The central control module is connected to the data acquisition and fusion module, the game model construction module, the dynamic inference and optimization module, the decision generation closed-loop module, the entropy calculation module, and the risk early warning and control module to coordinate the work of each module and the data interaction.

[0030] The steps for implementing the corresponding functions of each parameter and each unit module in the supply chain evolution game decision optimization system integrating multi-source data of the present invention can be referred to the parameters and steps in the embodiments of the supply chain evolution game decision optimization method integrating multi-source data mentioned above, and will not be repeated here.

[0031] Example 3 Based on the same inventive concept as Embodiment 1, this embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the aforementioned supply chain evolution game decision optimization method that integrates multi-source data by calling the computer program stored in the memory.

[0032] It should be noted that all computer programs for a supply chain evolution game decision optimization method that integrates multi-source data are implemented in C language.

[0033] Example 4 Based on the same inventive concept as in Embodiment 1, this embodiment proposes a computer-readable storage medium having an erasable and rewritable computer program stored thereon. When a computer program runs on a computer device, it enables the computer device to execute the aforementioned supply chain evolution game decision optimization method that integrates multi-source data.

[0034] For example, computer-readable storage media can be read-only memory, random access memory, read-only optical disc, magnetic tape, floppy disk, and optical data storage devices.

[0035] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. In particular, the embodiments for IoT devices and media are relatively simple in description because they are fundamentally similar to the method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0036] The systems, media, and methods provided in the embodiments of the present invention are in one-to-one correspondence. Therefore, the systems and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be repeated here.

[0037] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0038] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0039] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0040] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0041] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0042] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0043] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0044] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. The above are merely embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A supply chain evolutionary game decision optimization method integrating multi-source data, characterized in that, Includes the following steps: S1: Collect multi-source heterogeneous data from the supply chain, and perform cleaning, noise reduction, standardization, and trust verification on the collected multi-source heterogeneous data to generate fused global trustworthy status data. S2: Based on the global credible state data output in step S1, identify the game players and their strategy space in the supply chain, construct an improved dynamic evolutionary game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function, and reward / penalty coefficients. S3: Based on the game model constructed in step S2, an improved replicating dynamic equation is introduced to solve the strategy evolution, find the evolutionary stable strategy, and iteratively correct the model parameters by combining real-time data streams; S4: Generate differentiated decision-making schemes based on the evolutionary stable strategy obtained in step S3, establish a decision-making effect evaluation index system, and generate feedback data by comparing the actual execution effect with the expected goal to drive the model to optimize closed-loop parameters; S5: Calculate the state entropy of the comprehensive game based on the strategy selection ratio of each subject and the external environment indicators in the global credible state data; The calculation of the comprehensive game state entropy Qt based on the strategy selection ratios of each agent and external environment indicators in the global credible state data specifically includes the following steps: S51: The strategy distribution entropy is calculated based on the strategy selection ratio of each subject. The formula for calculating the strategy distribution entropy is: ； Where Ht is the strategy distribution entropy at time t, which measures the degree of dispersion of strategy choices among various entities in the supply chain. A larger value indicates a more uniform and chaotic strategy selection, while a smaller value indicates a more concentrated and rigid strategy. Let be the proportion of players in group i who choose strategy a at time t, N be the total number of players in the supply chain, and K be the total number of strategies that a single player can choose. This represents taking the logarithm of the policy proportion, used to quantify information uncertainty; S52: The external disturbance factor is calculated based on the external environment indicators in the global trusted state data. The formula for calculating the external disturbance factor is as follows: ； Where Dt is the external disturbance factor at time t, used to measure the overall volatility of the external environment indicators of the supply chain; the larger the value, the more severe the environmental disturbance. Let pj be the real-time value of the j-th external environmental indicator, zj be the historical mean of the j-th external environmental indicator, zj be the historical standard deviation of the j-th external environmental indicator, and M be the number of external indicators. S53: Calculate the state entropy of the comprehensive game based on the calculation results of the strategy distribution entropy and the calculation results of the external disturbance factor; S6: Based on the calculation results of the state entropy of the comprehensive game, a game imbalance early warning threshold is set to predict potential risks. When the early warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain.

2. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The trusted verification process in step S1 specifically includes: for multi-source heterogeneous data from the same supply chain, calculating the historical accuracy of each data source as the confidence weight, and generating a fused trusted data value through a weighted average method.

3. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The adaptive initialization formula for the profit function in step S2 is: ； in, Let be the adaptive payoff of agents i and j at time t when they adopt the strategy combination (a, b). This is the core decision-making basis for agents choosing strategies in evolutionary game theory, and it updates dynamically with time and the environment. t is a time variable, representing different moments in the dynamic evolution of the supply chain, reflecting the real-time nature of the payoff. is the basic revenue constant for the strategy combination (a, b), which is the benchmark revenue without considering demand fluctuations, costs, and synergistic rewards and penalties. It is pre-set by the characteristics of the supply chain business. y is the demand impact coefficient, used to adjust the strength of the impact of real-time demand fluctuations on the basic revenue. The real-time cost of agent i at time t when it adopts strategy a includes dynamic costs such as production costs, inventory costs, and logistics costs that change with time and strategy. It is calculated directly from multi-source fusion data in real time. i is the game agent number representing a participant in the supply chain, such as a supplier, manufacturer, or distributor. a is the strategy number adopted by agent i, such as high inventory strategy, low inventory strategy, or collaborative supply strategy. The reward / penalty coefficient between subjects i and j at time t is dynamically calculated based on the real-time market supply and demand ratio. When supply exceeds demand, the penalty for non-cooperative strategies is lowered; when supply is less than demand, the reward for cooperative strategies is raised. Let be the base value of the synergistic benefit of the strategy combination (a, b). It represents the fixed additional benefit or loss benchmark generated by cooperation such as information sharing and joint replenishment when subjects i and j adopt strategy (a, b). j is the number of another subject playing against subject i, representing a partner or competitor in the supply chain. b is the strategy number adopted by subject j, which together with subject i's strategy a constitutes the strategy combination (a, b).

4. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The improved replication dynamic equation formula in step S3 is as follows: ； in, The proportion of the main group i that chooses strategy a. To determine the expected return of strategy a, The average expected return of the population is given by numerically solving the replication dynamic equation until convergence, thus obtaining the evolutionary stable policy, where dt is a small time increment.

5. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The decision-making effectiveness evaluation index system in step S4 includes on-time delivery rate, total supply chain operating cost, and inventory level volatility index. By comprehensively comparing the deviation between the actual execution effect and the expected goal, the game model parameters in step S2 are triggered to optimize and adjust in a closed loop.

6. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The formula for calculating the state entropy of a comprehensive game is: ； Where Qt is the comprehensive game state entropy at time t, which is used to simultaneously characterize the comprehensive risk level of internal strategy structure and external environmental disturbances. It is the core basis for risk warning. α is the adjustment coefficient, which is used to control the influence weight of external disturbances on the comprehensive entropy. The larger the value of α, the stronger the influence of external disturbance factors on the system state.

7. The supply chain evolution game decision optimization method integrating multi-source data according to claim 1, characterized in that: The calculation results based on the comprehensive game state entropy are used to set a game imbalance early warning threshold to predict potential risks. When the early warning is triggered, a control strategy is automatically generated to intervene in the game process and ensure the stable operation of the supply chain. This includes the following steps: S61. Set the lower threshold Qmin and the upper threshold Qmax for the state entropy of the comprehensive game. S62. Risk assessment is based on the relationship between the state entropy Qt of the comprehensive game and the threshold. This indicates that the current supply chain faces a risk of strategic rigidity. This indicates that there is a risk of strategic confusion in the current supply chain; S63. Implement corresponding dynamic adjustment strategies based on different risk types. When the risk of strategy chaos is triggered, execute the first adjustment strategy to temporarily increase the collaborative return term in the return function. When the risk of policy rigidity is triggered, a second control policy is executed, introducing a random perturbation term δ into the replication dynamic equation, resulting in an improved replication dynamic equation: ； Wherein, δ is a random perturbation term used to break policy fixation and explore new policy spaces; S64. Continuously monitor the state entropy Qt of the comprehensive game until it returns to the safe range, and then gradually withdraw the control measures to restore the normal operation of the system.

8. A supply chain evolution game decision optimization system integrating multi-source data, implemented based on the supply chain evolution game decision optimization method integrating multi-source data as described in any one of claims 1-7, characterized in that, Specifically, it includes: a data acquisition and fusion module, a game model construction module, a dynamic deduction and optimization module, a decision generation closed-loop module, an entropy calculation module, a risk early warning and control module, and a central control module; The data acquisition and fusion module is used to collect multi-source heterogeneous data from the supply chain, and to clean, reduce noise, standardize and verify the collected multi-source heterogeneous data to generate fused global reliable status data. The game model construction module is used to identify the game subjects and their strategy space in the supply chain based on the global credible state data output in step S1, construct an improved dynamic evolution game model, and use real-time fused data to complete the adaptive initialization of the payoff function, cost function and reward / penalty coefficients. The dynamic deduction and optimization module is used to solve the strategy evolution problem based on the game model constructed in step S2. It introduces an improved replication dynamic equation to find an evolution-stable strategy and iteratively corrects the model parameters by combining real-time data streams. The decision generation closed-loop module is used to execute step S4 to complete the decision generation, evaluation and closed-loop optimization functions; The entropy calculation module is used to perform the comprehensive game state entropy calculation function described in step S5; The risk warning and control module is used to perform risk warning and dynamic control functions in step S6; The central control module is connected to the data acquisition and fusion module, the game model construction module, the dynamic inference and optimization module, the decision generation closed-loop module, the entropy calculation module, and the risk early warning and control module to coordinate the work of each module and the data interaction.

9. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor, characterized in that: the processor executes a supply chain evolution game decision optimization method that integrates multi-source data as described in any one of claims 1-7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: The system stores instructions that, when executed on a computer, cause the computer to perform a supply chain evolution game decision optimization method that integrates multi-source data as described in any one of claims 1-7.

Images