Nourishing dry cargo intelligent supply chain management system based on AI

Abstract

The invention discloses an AI-based nourishing dry goods intelligent supply chain management system, and belongs to the technical field of data processing and supply chain management, and the system comprises a data collection unit which is used for collecting historical sales data, weather forecast data, real-time road conditions, cold chain sensor data and market information; the AI prediction unit is used for carrying out demand and logistics path prediction based on the data acquired by the data acquisition unit so as to generate a suggested adjustment amount; the game balancing unit is used for carrying out multi-main-body game balancing analysis based on the data acquired by the data acquisition unit and a preset utility function so as to solve an elastic buffer threshold value; the decision execution unit is used for comparing the suggested adjustment amount generated by the AI prediction unit with the elastic buffer threshold calculated by the game balancing unit; the cohesion and the stability of the supply chain alliance are remarkably enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing and supply chain management, in particular to an AI-based intelligent supply chain management system for tonifying dry goods. BACKGROUND

[0002] In the supply chain management of high-end tonifying dry goods such as bird's nest, sea cucumber, fish maw, donkey-hide gelatin, ginseng, longan, and dendrobium, especially for rare products such as wild dendrobium and ginseng, artificial intelligence is widely used for demand prediction and logistics path planning; at the same time, cold chain technology is used to ensure the quality of goods during transportation; however, there is an inherent technical problem when existing technologies integrate AI prediction and cold chain logistics; when facing external strong disturbances such as extreme weather events, the AI system will frequently adjust the model and path in pursuit of real-time accuracy of prediction; this high-frequency decision-making change, i.e. decision oscillation effect, directly impacts the stability of the cold chain logistics transportation plan, which may lead to repeated scheduling of transportation tools, increased waiting time, and unexpected increase in the number of transshipment; this in turn increases the energy consumption of the cold chain and the loss rate of goods, ultimately reducing the overall efficiency and economic benefits of the supply chain. Existing solutions often focus on improving the performance of a single link and fail to solve the technical contradiction between dynamic optimization and stable operation.

[0003] To solve the above technical problems, the present application provides an AI-based intelligent supply chain management system for tonifying dry goods. SUMMARY

[0004] The present application aims to provide an AI-based intelligent supply chain management system for tonifying dry goods to solve the problems raised in the background art.

[0005] The technical solution of the present application is as follows:

[0006] A data acquisition unit for acquiring historical sales data, weather forecast data, real-time traffic, cold chain sensor data, and market intelligence;

[0007] An AI prediction unit for predicting demand and logistics paths based on the data acquired by the data acquisition unit to generate a recommended adjustment amount;

[0008] A game equilibrium unit for performing multi-agent game equilibrium analysis based on the data acquired by the data acquisition unit and a preset utility function to calculate an elastic buffer threshold;

[0009] A decision execution unit for comparing the recommended adjustment amount generated by the AI prediction unit with the elastic buffer threshold calculated by the game equilibrium unit; when the recommended adjustment amount exceeds the elastic buffer threshold, a direct adjustment signal is generated; when the recommended adjustment amount does not exceed the elastic buffer threshold, a delayed decision signal is generated;

[0010] The probability decision unit is used to perform probability evolution calculation on multiple candidate decision schemes in response to the delayed decision signal to generate the posterior probability of each scheme, and compare and analyze the posterior probability with the preset collapse threshold to generate a deterministic decision instruction.

[0011] Preferably, the game equilibrium analysis process is as follows: constructing utility functions for supplier entities, logistics entities and retailer entities in the supply chain; finding a Nash equilibrium strategy combination that enables all entities to achieve local optimal utilities at the same time through iterative calculations; extracting the optimal strategy adjustment amount of each entity in the Nash equilibrium strategy combination; and setting the minimum value of the optimal strategy adjustment amount as the elastic buffer threshold.

[0012] Preferably, the utility function components are measured in a unified monetary unit;

[0013] The utility function includes individual benefit function, operation cost function and penalty function;

[0014] The individual profit function is derived from the profit distribution model preset in the smart contract;

[0015] Operational cost function, data comes from the company's own financial and operational database;

[0016] The penalty function takes effect when the current policy deviates from the committed policy published on the blockchain.

[0017] Preferably, the penalty function includes an adjustment weight for the default penalty associated with reputation;

[0018] Adjust the weight and dynamically associate it with the subject’s reputation points recorded in the blockchain consensus layer;

[0019] The updating process of the subject's credit score is as follows: obtain the evaluation of the subject's performance behavior, and combine the current credit score with the preset learning rate to perform iterative updates; the evaluation of performance behavior is automatically executed by the smart contract based on whether the penalty function is triggered.

[0020] Preferably, the probability evolution calculation process is as follows:

[0021] The probability of the decision solution at the previous moment is obtained as the prior probability, and the new evidence currently collected is obtained; combined with the likelihood function that represents the probability of observing new evidence under the conditions for the decision solution to be established, Bayesian inference calculation is performed to obtain the posterior probability;

[0022] The likelihood function is constructed based on statistical analysis of historical data.

[0023] Preferably, the probabilistic decision unit is also used to: normalize the posterior probabilities of all candidate decision options; when the posterior probability of any decision option exceeds a preset collapse threshold, set the decision option as a deterministic decision instruction and issue it for execution.

[0024] Preferably, the decision execution unit is further configured to: in response to a direct adjustment signal, instruct the system to bypass the probability decision unit and directly execute an adjustment plan corresponding to the recommended adjustment amount.

[0025] Preferably, it operates in a multi-timescale layered framework, the layered framework including a strategic layer, a tactical layer and an operational layer;

[0026] The strategic layer is used to set long-term supply and demand targets and game frameworks on a monthly time scale;

[0027] The tactical layer is used to plan major logistics routes and inventory strategies on a weekly time scale;

[0028] The operation layer is used to perform the equilibrium analysis output by the game equilibrium unit and the probabilistic decision output by the probabilistic decision unit on a daily time scale.

[0029] The present invention provides an AI-based intelligent supply chain management system for nutritious dry goods through improvements. Compared with the existing technology, it has the following improvements and advantages:

[0030] 1. This solution unifies the conflicting interests of all parties within a computable mathematical framework by constructing a utility function for each entity and solving for Nash equilibrium strategy combinations. The elastic buffer threshold calculated by the game equilibrium unit is essentially the greatest common denominator recognized by all participants and reflects the global balance of interests. This move transforms the supply chain from a zero-sum game or information-asymmetric adversarial environment to a collaborative ecosystem seeking win-win results, significantly enhancing the cohesion and stability of the supply chain alliance.

[0031] 2. This solution quantifies the intangible asset of business reputation and directly impacts the economic benefits of all parties. Entities with high reputations enjoy lower penalty weights for breach of contract in the game, which is equivalent to a higher tolerance for decision-making errors. This provides an economic incentive for long-term honest behavior. Conversely, dishonest behavior will directly put them at a disadvantage in subsequent games. This design transcends traditional static contractual constraints and establishes a trust system that is self-regulating and promotes a virtuous cycle, with far-reaching value for maintaining long-term, stable supply chain partnerships.

[0032] 3. This scheme is not simply ignored, but the probability decision unit is enabled for careful processing; the unit is based on Bayesian inference, and the decision-making process is changed from a transient, deterministic judgment to a dynamic, evidence-driven belief updating process; the system can continuously collect new data as evidence and continuously correct the posterior probability of each candidate decision scheme; only when the credibility of a certain scheme exceeds the preset collapse threshold, the decision is finally determined; this delayed collapse mechanism effectively avoids the system making hasty and wrong decisions based on early, incomplete and noisy information, greatly improving the decision quality and reliability of the system in complex and variable market environments. BRIEF DESCRIPTION OF DRAWINGS

[0033] The application will be further explained below in conjunction with the accompanying drawings and embodiments:

[0034] Figure 1 is a flowchart of an AI-based nourishing dry goods intelligent supply chain management system of the application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical scheme and advantages of the application clearer, the application will be further described in detail below in conjunction with specific embodiments.

[0036] Embodiment 1:

[0037] Please refer to Figure 1 The application provides an AI-based nourishing dry goods intelligent supply chain management system, which comprises:

[0038] A data acquisition unit is configured to acquire historical sales data, weather forecast data, real-time traffic, cold chain sensor data and market intelligence.

[0039] An AI prediction unit is configured to perform demand and logistics path prediction based on the data acquired by the data acquisition unit, so as to generate a recommended adjustment amount.

[0040] A game equilibrium unit is configured to perform multi-agent game equilibrium analysis based on the data acquired by the data acquisition unit and a preset utility function, so as to solve an elastic buffer threshold.

[0041] A decision execution unit is configured to compare and process the recommended adjustment amount generated by the AI prediction unit and the elastic buffer threshold solved by the game equilibrium unit; when the recommended adjustment amount exceeds the elastic buffer threshold, a direct adjustment signal is generated; when the recommended adjustment amount does not exceed the elastic buffer threshold, a delayed decision signal is generated.

[0042] A probability decision unit is configured to respond to the delayed decision signal, perform probability evolution calculation on a plurality of candidate decision schemes, generate posterior probabilities of the schemes, and compare and analyze the posterior probabilities with a preset collapse threshold, so as to generate a deterministic decision instruction.

[0043] An AI-based intelligent supply chain management system for nutritious dry goods aims to resolve the technical contradiction between high-frequency AI prediction adjustments caused by external disturbances and stable physical operations. By introducing elastic buffers and probabilistic decision-making mechanisms, it achieves a balance between dynamic supply chain adaptability and operational stability. In this embodiment, the system is deployed in a distributed computing environment consisting of cloud servers and edge computing nodes.

[0044] The system's overall technical process begins with data collection, followed by AI prediction. The prediction results are not directly executed, but are filtered through an elastic buffer determined by game equilibrium analysis. Based on the comparison results, the system either directly executes or initiates a probabilistic decision-making process, ultimately outputting stable and multi-party consensus decision instructions.

[0045] The data collection unit aims to provide comprehensive, multi-dimensional data input for subsequent intelligent decision-making. The unit is an integrated data interface platform that aggregates data from various sources through APIs (application programming interfaces). Historical sales data comes from the company's internal ERP (enterprise resource planning) or CRM (customer relationship management system) to reveal the cyclicality and trends of demand. Weather forecast data comes from the national meteorological service agency to warn of extreme weather that may affect logistics. Real-time road conditions are obtained by calling the API of third-party map services to assess the immediate efficiency of transportation routes. Cold chain sensor data is uploaded in real time by IoT devices deployed in transportation vehicles and warehouses to monitor the status of goods. Market intelligence is captured from industry websites, social media and other channels using web crawler technology to capture sudden events and competitive product trends. All collected data is formatted and stored in a unified time series database.

[0046] The AI ​​prediction unit aims to make quantitative predictions of future market demand and optimal logistics routes based on the collected data. In this embodiment, the unit is composed of two deep learning models. The demand forecast adopts LSTM, a long short-term memory network model. In this embodiment, the LSTM model specifically adopts a stacked LSTM structure with two hidden layers, each with 128 neurons. To further improve the prediction accuracy, an attention mechanism is also integrated on the top of the model. The mechanism can dynamically assign weights to different time steps in the input sequence. For example, when predicting holiday demand, it can automatically give higher priority to holiday sales data from the same period of previous years. The model is pre-trained using three years of historical sales data, meteorological data, and market intelligence data to capture complex nonlinear relationships and time dependencies, generating daily demand forecasts for the future. Logistics route prediction uses a graph neural network model, which abstracts the logistics network into a graph structure where nodes represent warehouses or distribution centers and edges represent transportation routes. Edge weights are dynamically determined by real-time road conditions, transportation costs, and estimated durations. When new data input is received, the AI ​​prediction unit outputs a specific recommended adjustment amount, which refers to the quantitative adjustment required to the current operating plan to respond to the forecast change.

[0047] The game equilibrium unit aims to simulate the interest game of various supply chain participants and calculate a system elasticity range acceptable to all key participants to absorb small disturbances. The core of the unit is an algorithm based on the Nash equilibrium concept, which transforms the implicit interests of each party into an explicit mathematical model. Based on the real-time data collected by the data acquisition unit and the preset utility functions of each party, the unit solves a elastic buffer threshold. The threshold represents the maximum plan adjustment that the system can tolerate without destroying the overall collaborative stability of the supply chain.

[0048] The decision execution unit serves as the decision arbitration hub between AI predictions and final decision instructions. The unit receives the recommended adjustment amount generated by the AI ​​prediction unit and the elastic buffer threshold calculated by the game equilibrium unit, and compares the two numerically. When the recommended adjustment amount exceeds the elastic buffer threshold, it means that a major event that requires an urgent response has occurred. At this time, the unit will generate a direct adjustment signal. When the recommended adjustment amount does not exceed the elastic buffer threshold, it indicates that the disturbance is within the elastic range that the system can tolerate. At this time, the unit generates a delayed decision signal to initiate the subsequent probabilistic decision process.

[0049] The probabilistic decision-making unit is designed to respond to delayed decision signals and handle small perturbations within the elastic range in a more prudent manner, avoiding a rigid, deterministic response to every tiny fluctuation. Designed as a probabilistic evolutionary computation engine based on Bayesian inference, the unit maintains the probability distribution of multiple candidate decision options. When new information is acquired by the data acquisition unit, the unit iteratively updates the posterior probability of each option based on this new evidence. Only when the posterior probability of a particular option grows high enough to exceed a preset collapse threshold will the unit generate a deterministic decision instruction and issue it to the supply chain execution system.

[0050] This embodiment, through the collaborative work of the above-mentioned units, constructs a closed-loop intelligent decision-making system. Instead of directly coupling AI prediction results with logistics execution decisions, an elastic buffer zone and probabilistic decision-making mechanism, determined by multi-agent game theory, are established between the two. This solution effectively decouples the high-frequency prediction updates of AI from the stable logistics execution in the physical world. While maintaining the system's high sensitivity to environmental changes, it significantly enhances operational stability, reduces ineffective scheduling and resource waste caused by decision-making fluctuations, and ultimately improves the overall efficiency and economic benefits of the supply chain.

[0051] Existing technologies in supply chain management typically apply AI predictions directly to logistics execution. This direct coupling mechanism can trigger decision-making oscillations when faced with persistent, minor disturbances in the external environment. This can manifest as frequent changes in transportation plans, waste of scheduling resources, and unexpected increases in operating costs.

[0052] The core difference of this solution lies in the innovative introduction of two core units between the AI ​​prediction unit and the decision-making execution unit: the game equilibrium unit and the probabilistic decision unit; the design achieves effective decoupling of the prediction layer and the execution layer; the system no longer responds rigidly to every prediction fine-tuning, but filters through a dynamic elastic buffer threshold determined by the interest game of multiple parties; the introduction of this mechanism enables the system to absorb and smooth out most high-frequency, low-amplitude disturbances, and respond quickly only to major events that exceed the tolerance limits of all parties; this processing method suppresses decision-making oscillations from a mechanistic perspective, ensures the stability of transportation planning and warehousing operations in the physical world, and directly brings about the beneficial effects of reduced cold chain energy consumption, reduced product loss rate and improved overall operational efficiency.

[0053] The game equilibrium analysis process is as follows: construct utility functions for suppliers, logistics providers, and retailers in the supply chain; iteratively calculate and find a Nash equilibrium strategy combination that simultaneously achieves local optimal utility for all entities; extract the optimal strategy adjustment for each entity in the Nash equilibrium strategy combination; and set the minimum value of the optimal strategy adjustment as the elastic buffer threshold.

[0054] In this embodiment, the game equilibrium analysis process is concretized; the core purpose is to provide a computable and reproducible method to transform the abstract multi-party game into a specific elastic buffer threshold that can guide practice.

[0055] Construct utility functions for the core participants in the supply chain, namely suppliers, logistics providers, and retailers. A utility function is a mathematical expression that quantifies the net benefit or satisfaction an entity obtains under a specific strategy choice.

[0056] The system uses iterative calculations to find a Nash equilibrium strategy combination that simultaneously achieves a local optimum for all agent utilities. A Nash equilibrium refers to a strategy combination in a game where unilaterally changing the strategy of any participant does not increase utility. In this embodiment, the Nash equilibrium solution is executed by an off-chain server, which deploys a gradient descent or simulated annealing algorithm to find a strategy combination that simultaneously achieves a local optimum for all agent utility functions. After the calculation is complete, the final elastic buffer threshold, a key result, is submitted to the blockchain smart contract for recording and public disclosure. Subsequent decision execution units will read this threshold from the smart contract to ensure its authority, transparency, and immutability.

[0057] After calculating the Nash equilibrium strategy combination, the system will extract the optimal strategy adjustment amount of each subject in the Nash equilibrium strategy combination; the optimal strategy adjustment amount refers to the strategy adjustment value corresponding to each subject that can maximize its own utility in the Nash equilibrium state;

[0058] Based on the optimal policy adjustments from each party, the system takes the minimum value and sets it as the elastic buffer threshold. The elastic buffer threshold is determined by the minimum value. For example, if it is calculated that the supplier can tolerate a 2-hour delay, the logistics provider can tolerate a 5-hour delay, and the retailer can only tolerate a 1-hour fluctuation, then the final elastic buffer threshold will be set to 1 hour.

[0059] The game equilibrium analysis process defined in this way ensures that the resulting elastic buffer threshold is the greatest common denominator acceptable to all key participants. By unifying and balancing the interests of all parties through a mathematical model, the threshold setting is no longer subjective, but based on a transparent, fair, and quantifiable game outcome. This greatly enhances the universality of decision-making and the feasibility of execution in a multi-party collaborative environment, ensuring the stability of the supply chain alliance.

[0060] The prior art often focuses on the performance improvement of a single link, such as maximizing prediction accuracy or minimizing single transportation cost, but ignores the negative impact of these local optimal solutions on other participants in the supply chain; the present scheme unifies the conflicting interests of all parties into a computable mathematical framework by constructing an utility function for each subject and solving the Nash equilibrium strategy combination; the elastic buffer threshold calculated by the game equilibrium unit is essentially the greatest common divisor recognized by all participants, which is the embodiment of global interest equilibrium; this will change the supply chain from a zero-sum game or asymmetric information confrontation environment to a win-win collaborative ecosystem, significantly enhancing the cohesion and stability of the supply chain alliance.

[0061] The utility function components are measured in unified currency units;

[0062] The utility function includes an individual income function, an operating cost function, and a penalty function;

[0063] The individual income function is derived from the income distribution model preset in the smart contract;

[0064] The operating cost function is derived from the enterprise's own financial and operational databases;

[0065] The penalty function takes effect when the current strategy deviates from the commitment strategy published on the blockchain;

[0066] The utility function is the cornerstone of the system's game theory; it unifies the economic goals of each participant, including maximizing income, operating constraints, minimizing cost, and contract spirit, reputation punishment, into a monetary dimension; the function is not derived by the system, but is constructed based on the real financial data of each party and the preset distribution rules, and its output is the direct basis for the Nash equilibrium solution by the game equilibrium unit;

[0067] In this embodiment, the composition of the utility function is specified in detail based on the above embodiment; the purpose is to ensure the calculability of the utility function and the unity of the dimensions of each component, so that the influence of different dimensions can be weighed in a unified framework;

[0068] The utility function components are measured in unified currency units; this means that whether it is sales revenue, transportation cost or default penalty, it will be converted into the same currency unit, so that direct algebraic addition and subtraction operations can be performed, ensuring the mathematical validity of the model;

[0069] To clarify its composition, the utility function includes an individual income function, an operating cost function, and a penalty function; these three components jointly determine the net utility of a subject;

[0070] In order to make the construction process of the utility function clearer and more operable, this embodiment defines each component as follows: Suppose any entity in the supply chain In adopting strategies For example, the net utility when choosing a specific transportation time, storage temperature, etc. is , the calculation formula is:

[0071]

[0072] in: :Any entity in the supply chain In adopting strategies The net utility when : Represents any entity in the supply chain; : The strategy adopted by agent i, for example, choosing a specific transportation time, storage temperature, etc.; : Operation cost function, representing the subject To implement the strategy the costs required; : Penalty function, used to punish behaviors that deviate from the commitment strategy; Individual payoff function, representing the strategy adopted by subject i Expected income at the time of

[0073] Individual payoff function : The function quantifies the subject The expected income of

[0074]

[0075] In the formula, Is the subject The profit distribution ratio is preset and cannot be tampered with in the smart contract; is the total value of the batch of goods in the market; Is a strategy The relevant expected loss rate function, for example, if the strategy Including the transportation time, the longer the transportation time, the higher the loss rate;

[0076] Operation cost function : Function represents the body To implement the strategy The required cost; the data comes from the enterprise financial database and can be modeled as:

[0077]

[0078] in, For fixed costs, such as equipment depreciation, For and strategy Directly related variable costs, such as fuel costs and cold chain energy consumption, are functions of strategic variables, such as transportation distance and refrigeration power.

[0079] Penalty function :The function is used to punish the behavior of deviating from the commitment; let the subject The commitment policy published on the blockchain is , then the penalty function can be a quadratic function related to the degree of deviation:

[0080]

[0081] in, is the adjustment weight associated with the subject's reputation score; It is the basic penalty coefficient set for the entity, representing its benchmark cost of default. Its dimension is monetary unit, representing the basic penalty amount per unit deviation. Is the current strategy and commitment strategies The square of the Euclidean distance between them is used to quantify the severity of the deviation, ensuring that small deviations are subject to small penalties, while major deviations will result in exponentially increasing penalty costs; to ensure the effectiveness of the Euclidean distance calculation, before substituting into the formula, the strategy vector and Each component with physical units must be dimensionless in advance; for example, each component can be normalized by dividing it by its corresponding standard value or allowed fluctuation range; after this process, It becomes a dimensionless pure number that represents the relative severity of the strategy deviation;

[0082] The purpose of the individual profit function is to quantify the economic returns that an entity can obtain from a specific supply chain activity. The function is derived from the profit distribution model preset in the smart contract. The preset profit distribution model refers to a rule that is agreed upon in advance by all parties in the supply chain and solidified in the form of code on the blockchain. The individual profit function calculates the expected profit of each entity under the rule based on the real-time value of goods and the estimated loss rate.

[0083] The operational cost function aims to quantify the cost of implementing its strategy; the data for this function comes from the company's own financial and operational databases;

[0084] The penalty function aims to impose economic constraints on deviations from commitments to maintain the contractual spirit of the supply chain. The function takes effect when the current strategy deviates from the commitment strategy published on the blockchain. The commitment strategy refers to the operational plan publicly published and mutually agreed upon by all parties on the blockchain in advance. When an entity's actual implementation strategy deviates from this commitment, the penalty function is triggered, and the corresponding penalty amount is deducted from its total utility.

[0085] Through the above-mentioned refined definition of the utility function, this embodiment constructs a highly structured and quantitative game model; it successfully unifies multiple factors such as revenue, cost, and reputation in business activities into a single mathematical framework measured in currency; this makes the trade-offs between the interests of all parties precise, transparent, and automatically calculable, providing a solid mathematical foundation for finding Nash equilibrium solutions that truly reflect the core demands of all parties, thereby improving the accuracy and fairness of game equilibrium analysis.

[0086] The penalty function contains the adjustment weight of the default penalty associated with reputation;

[0087] Adjust the weight and dynamically associate it with the subject’s reputation points recorded in the blockchain consensus layer;

[0088] The process of updating the subject's credit score is as follows: an evaluation of the subject's performance is obtained and iteratively updated based on the current credit score and the preset learning rate. The evaluation of performance is automatically executed by the smart contract based on whether the penalty function is triggered.

[0089] This embodiment further optimizes the internal mechanism of the penalty function based on the above embodiment. The core purpose is to introduce a dynamic reputation mechanism that is linked to the subject's historical behavior. This makes the penalty no longer static but self-adjustable, thereby more effectively incentivizing long-term cooperation and honest performance of contracts.

[0090] The penalty function includes an adjustment weight for the default penalty associated with creditworthiness; the adjustment weight is a coefficient used to amplify or reduce the severity of the basic penalty;

[0091] Adjust the weights and dynamically associate them with the subject's reputation score recorded in the blockchain consensus layer. The subject's reputation score is a numerical value stored on the blockchain that quantifies the subject's historical performance. The immutable nature of the blockchain consensus layer ensures the credibility of the reputation score.

[0092] The update process of the subject's credit score is designed as an iterative learning process. The system obtains the evaluation of the subject's performance and combines the current credit score with the preset learning rate to perform iterative updates. The specific mathematical model of the process is as follows:

[0093]

[0094] in, For the theme New credit points at the next moment; : refers to subject i; : represents the next moment; : represents the current moment; : The evaluation of the subject's performance at the current time t is a binary evaluation result, for example, performance is 1 and breach of contract is 0; For the theme The reputation score at the current moment is derived from the historical records on the blockchain, and the value range is limited to a certain range, such as within the range of The learning rate is a constant between 0 and 1. The value is determined by analyzing the convergence and response speed of historical data and balancing stability and sensitivity.

[0095] Learning rate The value of can be optimized and determined by backtesting the historical performance data of the supply chain over the past 1-2 years; the goal of backtesting is to find an optimal The value of , makes the overall game framework composed of the reputation system and the penalty mechanism produce the lowest total default cost of the supply chain under the learning rate; the process can be formalized as an optimization problem:

[0096]

[0097] in, : The optimal learning rate λ value; : Indicates the operation of finding the parameter value that minimizes the total penalty cost of the objective function; : time point, from 1 to T; is the time length of historical data, It's at the time , the parameters are The total penalty cost generated by the reputation model; by simulating different The historical evolution of the value can choose the one that minimizes the total penalty cost. As the final configuration parameters of the system; : The state of the reputation model with λ as parameter at time point t;

[0098] The evaluation of contract performance is a binary evaluation result of the subject's behavior;

[0099] The evaluation of the performance of the contract is automatically executed by the smart contract according to whether the penalty function is triggered; the smart contract is deployed on the blockchain, and when it detects that the behavior of the subject deviates from the commitment strategy, causing the penalty function to be triggered, the smart contract automatically assigns the value of 0 to the current performance of the subject corresponding to the subject; otherwise, it is assigned a value of 1; this process is fully automated, avoiding human intervention;

[0100] The embodiment introduces a penalty weight dynamically associated with the reputation points to establish an adaptive incentive and constraint mechanism; for a subject with high reputation, the adjustment weight of the default penalty will be reduced accordingly, which is an incentive for long-term honest behavior; on the contrary, a subject with low reputation will face more severe punishment; this design not only effectively deters opportunistic behavior, but also encourages all participants to focus on long-term cooperation and actively maintain their reputation, thereby greatly enhancing the cohesion and long-term stability of the entire supply chain alliance, forming a virtuous cycle of trust ecology;

[0101] The iterative model of reputation points is an exponential moving average model, which describes how a subject's reputation evolves with its behavior; the future reputation of a subject is a weighted average of its historical reputation and the latest behavior performance; the learning rate determines the memory length of the system; a smaller value means that the system places more emphasis on long-term consistent performance, while a larger value makes the reputation more sensitive to recent behavior;

[0102] By introducing a penalty function dynamically associated with the reputation points of the subject, the scheme quantifies the intangible asset of business reputation and directly affects the economic utility of each party; a subject with high reputation enjoys a lower default penalty weight in the game, which is equivalent to a higher decision fault tolerance rate, which is an economic incentive for long-term honest behavior; on the contrary, dishonest behavior will directly lead to its disadvantaged position in subsequent games; this design goes beyond traditional static contract constraints and builds a trust system that can self-regulate and promote a virtuous cycle, which has far-reaching value in maintaining long-term and stable supply chain cooperation.

[0103] Embodiment 2

[0104] The probability evolution calculation process is as follows:

[0105] The probability of the decision scheme at the previous time is obtained as the prior probability, and the newly collected evidence is obtained; the likelihood function representing the probability of observing new evidence under the condition that the decision scheme is established is combined to perform Bayesian inference calculation to obtain the posterior probability;

[0106] The likelihood function is constructed based on statistical analysis of historical data;

[0107] In this embodiment, the probability evolution calculation process is specifically described; the purpose is to provide a rigorous mathematical method based on Bayesian inference for dynamically updating the credibility of each decision-making scheme under incomplete information conditions, thereby achieving prudent delayed decision-making.

[0108] The core logic of the process is to sequentially update the probability of decision options. The initial step is that the system obtains the probability of the decision option at the previous moment as the prior probability and obtains the new evidence currently collected. The prior probability refers to the initial judgment of the possibility of a decision option being valid before obtaining new information, and the value is derived from the results of the previous calculation cycle. The new evidence refers to the latest information related to the decision obtained by the data acquisition unit at the current moment.

[0109] The system performs Bayesian inference calculations based on the likelihood function that represents the probability of observing new evidence given the decision solution is valid, to obtain the posterior probability. The posterior probability refers to the probability that the updated decision solution is valid after considering the new evidence. The calculation follows the Bayesian formula:

[0110]

[0111] in, is the posterior probability; is a normalization constant, which ensures that the sum of the posterior probabilities of all candidate decision options is equal to 1; Is the likelihood function, which is used to measure the likelihood of decision If established, new evidence is observed probability; is the prior probability; : A decision-making plan to be selected; : new evidence collected at the current time t; : the specific new evidence observed at the current time t; : The evidence set of the previous moment;

[0112] The likelihood function is constructed based on statistical analysis of historical data;

[0113] In this embodiment, the likelihood function model is implemented by constructing a Gaussian mixture model; the system extracts a large number of decision-making solutions and subsequent evidence data pairs from the historical database and trains the GMM based on these data; after the training is completed, for any given decision-making solution and new observational evidence , GMM can calculate the probability density of observed evidence under the condition that the decision plan is established, that is, the likelihood value ;

[0114] The preset likelihood function model refers to a probability model obtained by mining massive historical data through machine learning methods; when new evidence emerges, the corresponding likelihood probability value can be queried from the model;

[0115] By adopting a probability evolution calculation process based on Bayesian inference, this embodiment makes the decision-making process no longer a black-and-white instantaneous judgment, but a dynamic, evidence-driven belief updating process; it allows the system to maintain the probability superposition state of the decision when information is insufficient, and gradually increases confidence in a certain optimal option as new information is continuously injected; this mechanism greatly improves the robustness and prudence of decision-making, and effectively avoids the risk of making hasty or erroneous decisions due to early, insufficient information.

[0116] The probabilistic decision unit is also used to: normalize the posterior probabilities of all candidate decision options; when the posterior probability of any decision option exceeds a preset collapse threshold, set the decision option as a deterministic decision instruction and issue it for execution;

[0117] In this embodiment, the function of the probabilistic decision unit is supplemented; the purpose is to define the clear triggering conditions for decision collapse, that is, how and when to transform from a probabilistic state with multiple possibilities to a unique deterministic instruction that must be executed;

[0118] The probabilistic decision unit is also used to normalize the posterior probabilities of all candidate decision options. Normalization is a mathematical operation that ensures that the sum of the posterior probabilities of all candidate decision options is exactly equal to 1. This step is a basic requirement of probability theory and ensures the validity of the probability distribution.

[0119] The core function of the unit is to judge and execute the collapse process of the decision; when the posterior probability of any decision scheme exceeds the preset collapse threshold, the decision scheme is set as a deterministic decision instruction and issued for execution; the collapse threshold, denoted as , is a preset value; the technical principle for setting the preset collapse threshold is based on risk cost assessment, and the value can be quantified by statistical analysis of the risk-return ratio of historical cases;

[0120] Collapse threshold The setting is based on the quantitative comparison of the decision action cost and the delay decision cost; for any candidate decision plan , the system evaluates the expected cost of executing it immediately The expected cost of waiting and gathering more information ;

[0121] Execution costs Including the direct operating costs of implementing the plan, as well as the risk costs caused by the plan being wrong; As its posterior probability, the expected cost is: ;

[0122] Cost of delay It refers to the potential loss of opportunity or accumulation of risk due to failure to act in a timely manner. The value can be calculated based on historical data statistics.

[0123] The system will continue to iterate until a solution is found. The posterior probability Make its expected execution cost lower than the delay cost for the first time, that is, satisfy:

[0124]

[0125] in, :Confirm the plan the cost of correct and successful execution; :Due to the plan Risk costs caused by errors; : The expected cost of delaying the decision and continuing to wait for information collection; by solving the critical probability in the above inequality , we can get the collapse threshold This approach ensures that decisions are triggered by clear economic trade-offs rather than subjective decisions.

[0126] When the posterior probability of a solution When the system believes that the evidence for the solution is sufficient and the uncertainty has been reduced to an acceptable level, the solution is collapsed from the probability superposition state and becomes a clear instruction that needs to be executed immediately.

[0127] The embodiment provides a clear termination and output mechanism for the probabilistic decision-making process; normalization ensures the logical integrity of probability calculations, while the introduction of a collapse threshold finds an adjustable balance between prudence and timeliness in decision-making. This enables the system to flexibly adjust the triggering conditions for its decision-making certainty based on the importance and risk level of the task, avoiding missed opportunities due to indecision and preventing rash actions due to insufficient evidence, significantly improving the intelligence and practicality of the entire decision-making system.

[0128] It improves the robustness of decision-making in uncertain environments; when faced with small disturbances within the elastic buffer zone, this solution does not simply ignore them, but instead uses a probabilistic decision-making unit to handle them prudently; based on Bayesian inference, the unit transforms the decision-making process from an instantaneous, deterministic judgment to a dynamic, evidence-driven belief updating process; the system can continuously collect new data as evidence and continuously revise the posterior probability of each candidate decision plan; the decision is finalized only when the credibility of a certain plan exceeds the preset collapse threshold; this delayed collapse mechanism effectively avoids the system from making hasty and erroneous decisions based on early, incomplete noise information, greatly improving the decision-making quality and reliability of the system in complex and changing market environments.

[0129] Example 3

[0130] The decision execution unit is further configured to: in response to the direct adjustment signal, instruct the system to bypass the probability decision unit and directly execute the adjustment plan corresponding to the recommended adjustment amount;

[0131] In this embodiment, the function of the decision execution unit is supplemented; the purpose is to provide the system with a bypass direct emergency response procedure to handle major disturbances that exceed the normal fluctuation range;

[0132] The decision execution unit is also used to: respond to a direct adjustment signal, instructing the system to bypass the probability decision unit and directly execute the adjustment plan corresponding to the recommended adjustment amount. The direct adjustment signal is an internal control signal generated by the decision execution unit itself when the recommended adjustment amount predicted by the AI ​​exceeds the elastic buffer threshold calculated by the game equilibrium. When the signal is triggered, it indicates that the external environment has changed drastically and the impact has exceeded the range determined by the multi-party game and can be smoothly absorbed by the system through delay and buffer mechanisms. In this case, the system logic will switch, and the recommended adjustment amount will no longer be sent to the probability decision unit. Instead, it will be directly converted into an adjustment plan and issued to the relevant execution system.

[0133] The dual-mode operation capability ensures that the system maintains operational economy through a buffer mechanism under conventional disturbances, and guarantees the resilience and timeliness of the supply chain through emergency response procedures when encountering major shocks, thereby greatly enhancing the overall survivability and adaptability of the system in complex and changing environments.

[0134] Example 4

[0135] It operates in a multi-timescale layered framework consisting of strategic, tactical, and operational layers.

[0136] The strategic layer is used to set long-term supply and demand targets and game frameworks on a monthly time scale;

[0137] The tactical layer is used for planning main logistics lines and inventory strategies on a weekly time scale;

[0138] The operational layer is used for executing the equilibrium analysis output by the game equilibrium unit and the probability decision output by the probability decision unit on a daily time scale;

[0139] In this embodiment, the entire system is arranged in a hierarchical framework operating on multiple time scales; the purpose is to decouple decision-making problems of different frequencies and different levels, so that the stability of high-level strategies and the flexibility of low-level operations can be considered; the framework includes a strategic layer, a tactical layer and an operational layer;

[0140] The strategic layer is used for setting long-term supply and demand targets and game frameworks on a monthly time scale; the decision frequency of the strategic layer is the lowest, and the macro direction is concerned; at this level, the manager will set the overall supply and demand balance target for the next month according to the annual sales target and market trend; more importantly, the core rules in the game framework, such as the constraint range and guiding principle for setting the basic penalty coefficient of each subject, are also defined and adjusted at this level;

[0141] The correlation model of the penalty weight establishes a direct negative correlation between reputation and economic punishment; the system significance is to convert abstract reputation points into actual influence in the game; the higher the reputation points of a subject, the smaller the cost of deviating from the commitment in the game, and vice versa; the model is a bridge connecting the reputation system and the game system;

[0142] The tactical layer is used for planning main logistics lines and inventory strategies on a weekly time scale; the decision of the tactical layer is connected with the upper and lower levels, and the frequency is moderate; according to the monthly target issued by the strategic layer, combined with the medium-term weather forecast and known information such as promotion activities, the main transportation network path for the next week is planned, and the inventory allocation strategy between the central warehouse and the regional warehouse is determined; in addition, the system parameters related to reputation, such as the learning rate of reputation update, are also configured at this level;

[0143] The operational layer is used for executing the equilibrium analysis output by the game equilibrium unit and the probability decision output by the probability decision unit on a daily time scale; the operational layer is the front line of decision execution, and the frequency is the highest; it completely operates according to the core logic described in the present application: every day or even every hour, the system continuously performs data collection, AI prediction, game equilibrium analysis, threshold comparison and possible probability decision; all daily, dynamic and high-frequency adjustments are completed at this level;

[0144] By constructing this hierarchical decision-making framework, this embodiment successfully decomposes the complex and large supply chain management problem into three distinct but interrelated sub-problems. The stability of the strategic layer provides clear direction and game rules for the entire system. The tactical layer translates strategic intentions into actionable weekly plans. The operational layer, through its unique elastic buffer and probabilistic decision-making mechanism, efficiently handles daily high-frequency disturbances while protecting tactical and strategic-level plans from these minor fluctuations. This hierarchical governance structure achieves a balance between macro-stability and micro-flexibility, greatly improving management efficiency and the overall robustness of the system.

[0145] The evolutionary model of probabilistic decision making is a direct application of Bayes' theorem in decision science. It describes the updating rule for a system's beliefs: the updated belief about a decision option, the posterior probability, is proportional to the initial belief, the product of the prior probability and the support for the option with new evidence, the likelihood function. The implementation of this formula enables the decision process to incorporate logical reasoning and evidence accumulation.

[0146] Through the organic combination of the above-mentioned mechanisms and models, this solution constructs an intelligent supply chain management system that can adapt to major environmental changes while resisting daily high-frequency disturbances. It achieves a delicate balance between dynamic adaptability and operational stability that cannot be achieved by existing technologies, and has significant technological progress and practical value.

[0147] 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 the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An AI-based intelligent supply chain management system for nutritious dry goods, characterized by: include: Data collection unit, used to collect historical sales data, weather forecast data, real-time traffic conditions, cold chain sensor data and market intelligence; An AI prediction unit, configured to perform demand and logistics route prediction based on the data collected by the data collection unit to generate a recommended adjustment amount; A game equilibrium unit, configured to perform a multi-agent game equilibrium analysis based on the data collected by the data collection unit and a preset utility function, so as to calculate an elastic buffer threshold; A decision execution unit is configured to compare the suggested adjustment amount generated by the AI ​​prediction unit with the elastic buffer threshold calculated by the game equilibrium unit; when the suggested adjustment amount exceeds the elastic buffer threshold, a direct adjustment signal is generated; when the suggested adjustment amount does not exceed the elastic buffer threshold, a delayed decision signal is generated; The probability decision unit is used to perform probability evolution calculation on multiple candidate decision schemes in response to the delayed decision signal to generate the posterior probability of each scheme, and compare and analyze the posterior probability with the preset collapse threshold to generate a deterministic decision instruction.

2. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 1 is characterized in that: The game equilibrium analysis process is as follows: construct utility functions for the supplier, logistics provider, and retailer entities in the supply chain; find a Nash equilibrium strategy combination that enables all entities to achieve local optimal utility simultaneously through iterative calculation; extract the optimal strategy adjustment amount of each entity in the Nash equilibrium strategy combination; and set the minimum value of the optimal strategy adjustment amount as the elastic buffer threshold.

3. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 2 is characterized in that: The components of the utility function are measured in a unified monetary unit; The utility function includes individual benefit function, operation cost function and penalty function; The individual profit function is derived from the profit distribution model preset in the smart contract; Operational cost function, data comes from the company's own financial and operational database; The penalty function takes effect when the current policy deviates from the committed policy published on the blockchain.

4. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 3 is characterized in that: The penalty function includes an adjustment weight of a default penalty associated with reputation; Adjust the weight and dynamically associate it with the subject’s reputation points recorded in the blockchain consensus layer; The updating process of the subject's credit score is as follows: obtain the evaluation of the subject's performance behavior, and combine the current credit score with the preset learning rate to perform iterative updates; the evaluation of performance behavior is automatically executed by the smart contract based on whether the penalty function is triggered.

5. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 1 is characterized in that: The probability evolution calculation process is as follows: The probability of the decision solution at the previous moment is obtained as the prior probability, and the new evidence currently collected is obtained; combined with the likelihood function that represents the probability of observing new evidence under the conditions for the decision solution to be established, Bayesian inference calculation is performed to obtain the posterior probability; The likelihood function is constructed based on statistical analysis of historical data.

6. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 5 is characterized in that: The probability decision unit is also used to: normalize the posterior probabilities of all candidate decision options; when the posterior probability of any decision option exceeds a preset collapse threshold, set the decision option as a deterministic decision instruction and issue it for execution.

7. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 1 is characterized in that: The decision execution unit is further configured to: in response to a direct adjustment signal, instruct the system to bypass the probability decision unit and directly execute an adjustment plan corresponding to the recommended adjustment amount.

8. The AI-based intelligent supply chain management system for nutritious dry goods according to claim 1 is characterized in that: Operates within a multi-timescale layered framework comprising strategic, tactical, and operational layers; The strategic layer is used to set long-term supply and demand targets and game frameworks on a monthly time scale; The tactical layer is used to plan major logistics routes and inventory strategies on a weekly time scale; The operation layer is used to perform the equilibrium analysis output by the game equilibrium unit and the probabilistic decision output by the probabilistic decision unit on a daily time scale.

Images