Supply chain management method and system based on intelligent manufacturing

Abstract

The invention relates to the technical field of supply chain management methods, in particular to a supply chain management method and system based on intelligent manufacturing, and the method comprises the steps: obtaining the cost information, operation and maintenance information and inventory information of each stage in a supply chain; based on the information, obtaining the risk value of each stage through a hierarchical weighted dynamic risk prediction algorithm; combining the risk value and the sales information of the commodity, and utilizing a superposition nonlinear gross profit optimization algorithm to obtain gross profit information of each stage; according to the gross profit information and the risk value, obtaining the optimal delivery distribution amount of each stage through an adaptive distribution optimization algorithm; based on gross profit accumulation of each stage, global gross profit optimization information is obtained through a multi-dimensional progressive gross profit accumulation optimization algorithm; based on global gross profit optimization information, feedback adjustment information of a supply chain system is formed through an overall feedback closed-loop optimization algorithm, and multi-dimensional risk prediction is formed by dynamically obtaining cost, operation and maintenance and inventory information of each stage of a supply chain.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management methods, and more specifically, to a supply chain management method and system based on intelligent manufacturing. Background Art

[0002] In the field of intelligent manufacturing supply chain management, with the increasing volatility of market demand and the complexity of the supply chain network, traditional supply chain management methods are gradually difficult to meet the refined and real-time management requirements. In the prior art, supply chain management usually relies on fixed cost, operation and maintenance, and inventory control methods, generally adopting simple linear analysis models or static prediction methods based on averages, lacking an agile response to dynamic market changes. This traditional approach has obvious limitations in supply chain optimization. Especially when facing real-time demand fluctuations, it is difficult to ensure the optimal state of revenue.

[0003] Currently, most supply chain management systems on the market are mainly optimized through two approaches: one is to reduce costs by improving the production efficiency of individual links, and the other is to perform a simple trend analysis on demand through a linear prediction model. However, these methods usually ignore the complex correlations and mutual influences between different stages, resulting in limited optimization capabilities for the overall supply chain. For example, a simple cost-based allocation model cannot dynamically consider the potential impact of risks at each link, which may lead to excessive inventory in some stages, while insufficient supply in other stages affects the overall revenue. Further, this model cannot fully reflect the real-time nature of market fluctuations. When there are large fluctuations in market demand, the system is difficult to quickly adjust the mismatch between demand and supply, resulting in damaged revenue.

[0004] In addition, the calculation of gross profit in traditional supply chain management mostly adopts a fixed ratio model, ignoring the potential optimization space of risk prediction and allocation strategies for gross profit, making it difficult to find a balance between revenue and risk. For example, gross profit is usually set as the difference between sales revenue and fixed costs. Although this fixed calculation method is easy to operate, it is difficult to adapt to demand changes at each stage and cannot fully consider risk differences under different market conditions. Therefore, the traditional gross profit calculation lacks flexibility and optimization depth, making it impossible to make full use of resources in the supply chain for flexible adjustment when market conditions fluctuate violently.

[0005] Existing supply chain management systems also have limitations in distribution strategies. Many systems adopt a single distribution plan or fixed rules in the distribution algorithm. This approach may be relatively effective when market demand is stable, but when demand fluctuates greatly, the distribution rules are prone to inventory backlogs or shortages due to their lack of adaptive adjustment capabilities. For example, in traditional distribution methods, the shipment volume is usually set based on historical averages, which can meet demand when market demand is stable, but once demand increases or decreases, the static distribution volume is likely to lead to uneven resource allocation.

[0006] In addition, there is generally a lack of a cumulative feedback mechanism for overall revenue in existing supply chain systems. Traditional methods mostly calculate costs and gross profits independently at each stage, lacking global feedback optimization for all stages of the supply chain. Therefore, in this case, the feedback between different stages cannot form a closed-loop control, resulting in a lack of coordination between the decisions of each stage of the supply chain and the inability to maximize overall revenue. In each link of the supply chain, due to the lack of interaction, it is difficult to optimize revenue cumulatively on a global scale, and thus it is impossible to achieve the effect of synergy and efficiency improvement. Summary of the Invention

[0007] Based on the above deficiencies of the existing technology, the present invention proposes a supply chain management method based on intelligent manufacturing, aiming to solve the problems of lack of dynamics, real-time performance, and coordination in existing methods through a series of innovative algorithms.

[0008] The present invention provides a supply chain management method for intelligent manufacturing, and the method includes:

[0009] Obtain the cost information, operation and maintenance information, and inventory information of each stage in the supply chain;

[0010] Based on the above information, obtain the risk value of each stage through a hierarchical weighted dynamic risk prediction algorithm;

[0011] Combine the risk value and the sales information of the product, and use a superimposed non-linear gross profit optimization algorithm to obtain the gross profit information of each stage;

[0012] According to the gross profit information and the risk value, obtain the optimal shipment allocation volume of each stage through an adaptive allocation optimization algorithm;

[0013] Based on the cumulative gross profit of each stage, obtain the global gross profit optimization information through a multi-dimensional progressive gross profit cumulative optimization algorithm;

[0014] Based on the global gross profit optimization information, form the feedback adjustment information of the supply chain system through an overall feedback closed-loop optimization algorithm.

[0015] Preferably, the hierarchical weighted dynamic risk prediction algorithm includes:

[0016] Generate a multi-dimensional risk factor vector V based on the cost information, operation and maintenance information, and inventory information at each stage i ; Use a weighted matrix W i to weight the risk factor vector to obtain a risk value R i ; The calculation formula for the risk value R i is as follows:

[0017]

[0018] where R i is the risk value at the current stage, W i is the risk prediction weighted matrix, V i is the multi-dimensional risk factor vector within the stage, Q i is the risk threshold vector, α k is the influence weight of the risk factor, β k is the risk adjustment coefficient, M i is the actual gross profit of the stage, K i is the expected value of gross profit prediction.

[0019] Preferably, the superposition non-linear gross profit optimization algorithm includes:

[0020] Based on the risk value R i and the cost information and operation and maintenance information of the commodity, determine the gross profit G at the current stage i ; The calculation formula for the gross profit G i is as follows:

[0021]

[0022] where G i is the stage gross profit, S i is the stage sales volume, C i is the stage cost, M i is the stage operation and maintenance cost, λ is the non-linear adjustment factor, R i is the risk value output of the previous layer, and θ is the risk balance point.

[0023] Preferably, the adaptive allocation optimization algorithm includes:

[0024] According to the gross profit G i and the risk value R i , obtain the shipping allocation quantity D through segmented adjustment i+1 ; The calculation formula for the shipping allocation quantity is:

[0025]

[0026] where D i+1 is the allocation quantity in the i + 1 stage, D iis the basic allocation quantity for the i-th stage, γ i is the risk allocation balance coefficient, δ is the risk threshold, and ζ is the adaptive allocation adjustment coefficient.

[0027] Preferably, the multi-dimensional progressive gross profit cumulative optimization algorithm includes:

[0028] Based on the gross profit information of each stage, combined with historical gross profit, perform progressive cumulative optimization of the gross profit to obtain the global gross profit G total ; The global gross profit G total The calculation formula is:

[0029]

[0030] Among them, G total is the cumulative gross profit, G i is the gross profit of the i-th stage, η i is the cumulative optimization factor of the stage, ξ is the gross profit growth sensitivity coefficient, and G i-1 is the gross profit of the previous stage.

[0031] Preferably, the overall feedback closed-loop optimization algorithm includes:

[0032] Based on the global gross profit, combined with the risk value and allocation information, form the feedback optimization quantity F i ; The feedback optimization quantity F i The calculation formula is:

[0033]

[0034] Among them, F i is the stage feedback optimization quantity, G total is the cumulative gross profit, R j is the risk value of the j-th stage, and ω j is the feedback adjustment coefficient.

[0035] The supply chain management system based on intelligent manufacturing that executes the method includes:

[0036] An information acquisition module for acquiring cost information, operation and maintenance information, and inventory information of each stage;

[0037] A risk prediction module that obtains the risk value of each stage based on the hierarchical weighted dynamic risk prediction algorithm;

[0038] A gross profit calculation module that obtains the gross profit information of each stage based on the superimposed non-linear gross profit optimization algorithm;

[0039] An allocation optimization module that obtains the shipment allocation quantity of each stage based on the adaptive allocation optimization algorithm;

[0040] The gross profit accumulation module obtains global gross profit information based on the multi-dimensional progressive gross profit accumulation optimization algorithm;

[0041] The closed-loop feedback module obtains feedback optimization information based on the overall feedback closed-loop optimization algorithm and updates it to the initial stage of the system.

[0042] Preferably, the risk prediction module obtains risk values at each stage based on the information obtained by the information acquisition module.

[0043] Preferably, the gross profit calculation module obtains gross profit information at each stage based on the risk values obtained by the risk prediction module.

[0044] Preferably, the closed-loop feedback module generates feedback optimization information based on the global gross profit information obtained by the gross profit accumulation module and updates it to the information acquisition module.

[0045] The invention has the following beneficial effects:

[0046] The supply chain management method and system of the present invention design algorithm modules such as hierarchical weighted dynamic risk prediction, superimposed non-linear gross profit optimization, adaptive allocation optimization, multi-dimensional progressive gross profit accumulation, and overall feedback closed-loop optimization to achieve dynamic prediction and optimization at each stage of the supply chain.

[0047] The present invention dynamically obtains cost, operation and maintenance, and inventory information at each stage of the supply chain to form multi-dimensional risk prediction. In particular, the hierarchical weighted dynamic risk prediction algorithm of the present invention can not only flexibly adapt to different market fluctuations, but also establish weights among various risk factors to achieve more accurate risk prediction. Based on this prediction result, the gross profit is refined and adjusted through the superimposed non-linear gross profit optimization algorithm to maximize the gross profit under the premise of controllable risk. The adaptive allocation optimization algorithm adjusts the shipment volume in real time, enabling the supply chain to still maintain reasonable allocation of resources during demand fluctuations and avoiding the risks of inventory backlog and shortage. The multi-dimensional progressive gross profit accumulation optimization algorithm dynamically accumulates the gross profit at each stage to achieve global optimization of the benefits. Finally, through the overall feedback closed-loop optimization algorithm, the system can form real-time feedback among each stage, continuously improving the overall efficiency of the supply chain.

[0048] Therefore, the technical problem to be solved by the present invention is how to achieve real-time optimization of supply chain management through accurate risk prediction, dynamic gross profit optimization, and adaptive allocation in the intelligent manufacturing environment, and achieve synergy in terms of global revenue accumulation and feedback regulation. Each step in the method of the present invention forms a coordinated and synergistic overall solution through the complementarity and superposition effect of the algorithms, effectively making up for the deficiencies of the prior art in dynamic regulation and real-time optimization, and ensuring the robustness and efficiency of the supply chain system under the condition of large market demand fluctuations. Description of the Drawings

[0049] Figure 1 This is the method flow logic block diagram of the present invention.

[0050] Figure 2 This is the system structure logic block diagram of the present invention.

[0051] Figure 3 This is the information flow and workflow judgment logic block diagram of the present invention. Detailed implementation manners

[0052] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following combines the accompanying drawings and preferred embodiments, and details its specific implementation manners, structures, features and effects as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0054] Please refer to Figures 1 - 3 , the present invention provides a supply chain management method and system based on intelligent manufacturing, aiming to achieve precise management of information such as costs, gross profits, and inventories at all stages of the supply chain through innovative algorithms of dynamic optimization, and perform allocation optimization based on risk prediction and gross profit analysis, so as to improve the overall efficiency of the supply chain in the intelligent manufacturing environment.

[0055] The core of the present invention lies in the supply chain management method based on intelligent manufacturing, and the key lies in first obtaining the cost information, operation and maintenance information, and inventory information at all stages of the supply chain. This information acquisition module can collect various types of information in real time from multiple data sources of the supply chain management system, such as the cost data at all stages recorded in the financial system, the equipment maintenance and operation costs in the operation and maintenance system, and the inventory data in the warehouse management system. Taking specific numerical values as examples, the cost information includes but is not limited to the unit cost of a single product in the production stage (such as 10 yuan to 15 yuan per piece), the operation and maintenance information includes the maintenance cost of each device (such as about 1 yuan to 5 yuan per day), and the inventory information represents the current inventory level. For example, the inventory at a certain stage may be between 200 pieces and 500 pieces.

[0056] Then, by summarizing and standardizing the data at all stages, the system can obtain the overall operating status of each node in the supply chain. Preferably, through the refined layering of data, the present invention can obtain the cost composition at different stages more precisely, thereby providing basic data for subsequent gross profit calculation and risk control, and ensuring the global optimization of supply chain management.

[0057] In a preferred embodiment of the present invention, based on information acquisition, the present invention further proposes a hierarchical weighted dynamic risk prediction algorithm to obtain the risk values at each stage. This algorithm uses a weighted matrix and a multi-dimensional risk factor vector for risk prediction. The specific formula is as follows:

[0058]

[0059] Where, R i is the risk value at the current stage, W i is the risk prediction weighted matrix, defined as an adjustable weight factor matrix, which can reflect the different impacts of various risk factors on the overall risk; V i is the multi-dimensional risk factor vector within the stage, used to represent the risk parameter Q of each stage i is the risk threshold vector, used to represent the reasonable range of each risk factor; α k is the influence weight of the risk factor, determining the weight ratio of each factor to the overall risk value; β k is the risk adjustment coefficient, usually adjusted according to historical data to optimize the prediction accuracy; M i is the actual gross profit at the current stage, and K i is the expected value of the gross profit prediction.

[0060] Preferably, in an embodiment of the present invention, the risk threshold can be set to 0.7, which is an empirical value, indicating that when the risk value exceeds 0.7, a certain stage of the supply chain is in a high-risk state and needs to enter the adjustment link. Based on this prediction algorithm, the system can flexibly adjust risk judgment based on real-time data, thus significantly improving the risk resistance ability and prediction accuracy of the supply chain.

[0061] On the basis of risk prediction, further by superimposing a non-linear gross profit optimization algorithm, the gross profit information at each stage is obtained. This algorithm realizes the optimal allocation of gross profit on the premise of controllable risk through the non-linear combination of risk value and gross profit. The specific calculation formula is as follows:

[0062]

[0063] Where, G i is the gross profit at the current stage, S i represents the stage sales volume, preferably set between 500,000 yuan and 1 million yuan in the case of large production and shipment volume to increase profits; C i is the stage cost, referring to the raw materials and production costs per unit of product, which can be 100,000 yuan to 200,000 yuan; M i is the operation and maintenance cost, usually about 50,000 yuan per stage; λ is the non-linear adjustment factor, used to control the smoothness of the gross profit curve, usually can be set from 0.5 to 1.5; R iis the risk value output by the previous algorithm, and θ is the risk balance point, which can usually be taken as 0.7 to avoid serious impact on gross profit due to excessive risk.

[0064] This algorithm preferably aims to reduce the fluctuation of gross profit in the high-risk stage. When the risk value reaches above 0.7, the algorithm limits the gross profit within a reasonable range by adjusting the proportional relationship between sales and cost, thereby reducing the volatility of system operation and maximizing the overall income of the supply chain.

[0065] In a preferred embodiment of the present invention, according to the optimized gross profit information and risk value, the optimal shipment allocation quantity for each stage is obtained through an adaptive allocation optimization algorithm. This algorithm is set by a piecewise function to dynamically adjust the allocation quantity for each stage according to the current risk and gross profit situation to ensure the optimal operation of the supply chain. The calculation formula is as follows:

[0066]

[0067] where D i+1 is the shipment allocation quantity for the i + 1 stage, and D i is the basic allocation quantity for the previous stage; γ i is the risk allocation balance coefficient, which can preferably be set within the range of 1.0 to 2.0 to balance the relationship between the shipment quantity and risk; δ is the risk threshold, and it is recommended to be 0.5 to 0.7 to adapt to different risk allocation strategies; ζ is the adaptive allocation adjustment coefficient, preferably between 0.3 and 0.7, to ensure that the system can quickly adapt to changes in a higher-risk environment.

[0068] This algorithm enables the supply chain to automatically optimize the shipment quantity allocation under different risks through a piecewise adjustment method. Preferably, when the risk value is lower than 0.5, the algorithm will maximize the allocation quantity to meet market demand; when the risk exceeds 0.7, the algorithm will significantly reduce the allocation quantity to reduce inventory backlog and cost waste, thereby enhancing the flexibility of the supply chain.

[0069] In a preferred embodiment of the present invention, based on the gross profit information of the foregoing stage, a multi-dimensional progressive gross profit accumulation optimization algorithm is used to perform the cumulative optimization of the gross profit for each stage to achieve the optimal allocation of the global gross profit of the supply chain. This algorithm combines historical gross profit and current stage data to maximize the global benefit by dynamically updating the cumulative gross profit. The specific formula is:

[0070]

[0071] where G total is the global cumulative gross profit, representing the overall income level of the current supply chain; G i is the gross profit value for the i stage; η iis the cumulative optimization factor for the stage, and the recommended range is 0.2 to 0.6 to adapt to different gross profit increase amplitudes; ξ is the gross profit growth sensitivity coefficient, usually 0.5 to 1.0, to adjust the sensitivity of gross profit growth; G i-1 is the gross profit of the previous stage.

[0072] Preferably, by gradually accumulating and adjusting the gross profit distribution, the supply chain can achieve optimal revenue distribution within different stages, while avoiding instability caused by excessive gross profit differences between stages. In addition, through the progressive optimization method, the gross profit of each stage can be maintained within a reasonable range, effectively improving the overall efficiency of the supply chain.

[0073] In a preferred embodiment of the present invention, based on the above-mentioned cumulative gross profit optimization, the present invention further provides an overall feedback closed-loop optimization algorithm to adjust the supply chain management strategy through real-time feedback, thereby optimizing the overall operation of the supply chain. This algorithm is based on the gross profit and risk data of each stage, and feeds back the global gross profit information to the starting stage of the system, enabling subsequent stages to adjust parameters according to the feedback, and then optimizing the overall efficiency and stability of the supply chain. The calculation formula of this algorithm is:

[0074]

[0075] where, F i is the feedback optimization amount for the current stage, representing the final optimization result of the i-th stage, which will be directly fed back to the next stage; G total is the cumulative gross profit, representing the overall revenue level of the supply chain; R j is the risk value of the j-th stage, used to represent the impact degree of this stage on the overall revenue; ω j is the feedback adjustment coefficient, preferably taking a value between 0.8 and 1.2 to ensure the rationality and stability of the feedback data.

[0076] In an embodiment of the present invention, the feedback adjustment coefficient can be adapted according to the specific risk value. For example, a lower adjustment coefficient is selected when the risk value is low (such as below 0.3), and the coefficient is increased when the risk value is high (such as above 0.7), so as to achieve the dynamic regulation effect under different risk conditions. Through this closed-loop feedback algorithm, the system can continuously optimize the relationship between gross profit and risk distribution, form an adaptive adjustment mechanism, and improve the overall response ability and profitability of the supply chain.

[0077] The present invention also provides a supply chain management system based on intelligent manufacturing, which includes an information acquisition module 1, a risk prediction module 2, a gross profit calculation module 3, an allocation optimization module 4, a gross profit accumulation module 5 and a closed-loop feedback module 6 to achieve dynamic monitoring and optimization of data at each stage. The information acquisition module 1 in this system is responsible for obtaining cost information, operation and maintenance information and inventory information from different data sources to ensure the comprehensiveness and accuracy of the information.

[0078] The risk prediction module 2 uses a hierarchical weighted dynamic risk prediction algorithm to analyze the risk factors at different stages and outputs the risk value to the gross profit calculation module 3. The gross profit calculation module 3 further uses a superimposed nonlinear gross profit optimization algorithm to calculate the gross profit information at each stage based on the comprehensive risk value. The allocation optimization module 4 uses an adaptive allocation optimization algorithm to allocate and adjust the gross profit results to achieve the optimal delivery allocation at each stage.

[0079] In addition, the gross profit accumulation module 5 combines the gross profit information of multiple stages, performs cumulative optimization through a multi-dimensional progressive gross profit accumulation optimization algorithm, and outputs the global gross profit information to the closed-loop feedback module 6. The closed-loop feedback module 6 uses an overall feedback closed-loop optimization algorithm to provide feedback results to the information acquisition module 1, thereby completing the closed-loop control of the entire system. Through the connection and coordinated operation of data flows between the modules of the system, a highly automated supply chain optimization system is formed.

[0080] In the supply chain management system based on intelligent manufacturing of the present invention, the risk prediction module 2 specifically applies the hierarchical weighted dynamic risk prediction algorithm. The algorithm dynamically predicts the risk value at each stage based on the data obtained by the information acquisition module 1. Its weighted matrix and risk factor vector can flexibly adapt to different risk types and influencing factors, thereby outputting accurate risk values.

[0081] In this process, the weighted matrix can be composed of different risk weight parameters to adapt to various market conditions. For example, in the case of drastic market fluctuations, the system can increase the weight of the market volatility parameter to more accurately reflect the impact of market changes on the supply chain. The output of the risk prediction module 2 directly affects the optimization decision of the gross profit calculation module 3, so that each stage of the supply chain can operate smoothly and obtain the best gross profit under the premise of controllable risks.

[0082] In the above system architecture, the gross profit calculation module 3 specifically adopts a superimposed nonlinear gross profit optimization algorithm to obtain accurate gross profit information based on the output value of the risk prediction module 2. Preferably, this module will compare historical data with real-time data, so as to adjust the λ and θ values in time when the parameter uncertainty is large to maintain the stability of the algorithm output.

[0083] For example, during the peak period of market demand, the change range of sales may be relatively large. At this time, the gross profit calculation module 3 can appropriately adjust the risk balance point \(\theta\) to ensure the adaptability of the algorithm to sales fluctuations. Through this non-linear optimization, the gross profit calculation module 3 can allocate supply chain resources more precisely and maximize the benefits.

[0084] In addition, the closed-loop feedback module 6 of the present invention feeds back the global gross profit and risk information to the information acquisition module 1 according to the overall feedback closed-loop optimization algorithm of claim 5. The core of the algorithm of the closed-loop feedback module 6 lies in the real-time optimization of the parameter settings of the information acquisition module 1 according to the overall revenue situation and risk level, so as to ensure that the supply chain can react quickly and optimize resources in the changing market environment.

[0085] In a preferred embodiment, the closed-loop feedback module 6 can set a feedback adjustment coefficient ω j between 0.8 and 1.2, so as to slightly reduce the resource input when the risk is relatively high, and increase the resource allocation in the low-risk stage, so as to enhance the balance of the supply chain. In this way, the feedback module can maintain the dynamic stability of the system in the changes of demand fluctuations and risk levels, and provide high-quality decision-making support for the subsequent stage.

[0086] Through the above detailed description, the supply chain management method and system based on intelligent manufacturing of the present invention form a complete logical closed-loop in aspects such as information acquisition, risk prediction, gross profit calculation, delivery allocation, global gross profit accumulation, and feedback adjustment. Through the collaborative action of algorithms and the real-time flow of data between modules, the system can accurately predict and control the risks and benefits of each stage of the supply chain, and finally realize the optimization of the overall efficiency of the supply chain.

[0087] The supply chain management method and system based on intelligent manufacturing of the present invention realize risk prediction, gross profit optimization, allocation adjustment, and feedback optimization of each stage of the supply chain through innovative algorithms to improve the overall efficiency and benefits of the supply chain. In order to verify the superiority of the present invention, experimental tests were carried out. By using a professional supply chain data set, a comparison was made between the method of the present invention and the existing traditional supply chain management method.

[0088] The data set used in the test covers multi-dimensional data of typical supply chain processes, including costs, operation and maintenance costs, inventory information, sales volume, and market fluctuations at each stage. The specific data is from the annual supply chain data of a certain intelligent manufacturing enterprise. This data set can reflect the changes in the supply chain under different market conditions, such as demand fluctuations between holiday peaks and off-seasons. For the convenience of comparison, the management method of the present invention (embodiment) and the traditional supply chain management method (comparative example) were respectively implemented under the same market conditions, and the gross profit, risk, delivery allocation accuracy, and overall revenue level at each stage were analyzed in detail.

[0089] The following four indicators are used in this test to verify the superiority of the method of the present invention in practical applications:

[0090] 1. Gross profit margin: Defined as the ratio of gross profit to sales, to verify the effect of the gross profit optimization algorithm. The detection method is to compare the sales and gross profit at each stage. The higher the gross profit margin, the better the revenue optimization effect.

[0091] 2. Risk control level: Quantified using risk values. The risk threshold set in the test is 0.7, and risks higher than this threshold are regarded as high risks. The detection of risk values is measured through the output of the risk prediction algorithm. The higher the risk control level, the stronger the risk regulation ability of the method.

[0092] 3. Allocation accuracy: Refers to the matching degree between the shipped allocation quantity and the actual demand. The allocation accuracy is detected by calculating the ratio of the actual shipment to the predicted demand at each stage. The higher the allocation accuracy, the better the adaptability and accuracy of the allocation algorithm.

[0093] 4. Overall revenue growth rate: Defined as the percentage increase in revenue of the management method of the present invention compared to the traditional method. The calculation method is (Overall revenue of the present invention - Overall revenue of the traditional method) / Overall revenue of the traditional method * 100%. This indicator is used to evaluate the improvement in overall benefits of the present invention.

[0094] In the embodiment, the method of the present invention performs real-time risk prediction, gross profit optimization, and dynamic allocation adjustment based on the information of each stage in the dataset; while the comparative method uses traditional average allocation and fixed gross profit calculation methods, without considering factors such as risk regulation and feedback optimization. The experimental results are shown in Table 1.

[0095] Index Example Comparative example Improvement rate Gross profit margin 45.6％ 38.2％ +7.4％ Risk control level 92％ 75％ +17％ Allocation accuracy 96％ 82％ +14％ Overall revenue growth rate 12.5％ - -

[0096] It can be seen from the detection results that the method of the present invention shows significant superiority in each indicator. In terms of the gross profit margin, the method of the present invention realizes higher revenue optimization by superimposing the non-linear gross profit optimization algorithm and combining real-time risk prediction and regulation, increasing the gross profit margin by 7.4 percentage points compared to the comparative method. This shows that the method of the present invention can utilize resources more effectively and maximize revenue.

[0097] In terms of the risk control level, through the combination of the hierarchical weighted dynamic risk prediction algorithm and the overall feedback closed-loop optimization algorithm, the risk regulation ability of the embodiment is significantly enhanced compared to the comparative method. In the comparative method, the risk control level is only 75%, while in the embodiment it reaches 92%, fully reflecting the stability and anti-risk ability of the present invention under complex market conditions.

[0098] The allocation accuracy is also one of the key advantages. Through the adaptive allocation optimization algorithm, the method of the present invention achieves a high degree of matching between the shipment volume in each stage and the actual demand. In the embodiment, the allocation accuracy reaches 96%, while that of the comparative example is only 82%. This shows that the present invention can accurately allocate the shipment volume under the condition of large fluctuations in actual demand, thereby reducing inventory costs and waste.

[0099] In terms of the overall revenue increase rate, the embodiment achieves a 12.5% revenue increase compared with the comparative example, indicating that under the same market conditions, the method of the present invention can significantly improve the overall efficiency through refined risk control and gross profit optimization, reaching a higher management level and economic benefit.

[0100] In summary, the test results prove the superiority of the supply chain management method and system based on intelligent manufacturing of the present invention in terms of improving the gross profit margin, risk control degree, allocation accuracy, and overall revenue increase rate. The innovative algorithm of the present invention performs better than the traditional method in each index, verifying the practical application value of the present invention in the intelligent manufacturing supply chain scenario.

[0101] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A supply chain management method based on intelligent manufacturing, characterized in that The method includes: Obtaining the cost information, operation and maintenance information, and inventory information at each stage of the supply chain; Based on the above information, obtaining the risk value at each stage through a hierarchical weighted dynamic risk prediction algorithm; Combining the risk value and the sales information of the commodity, and obtaining the gross profit information at each stage by using a superimposed non-linear gross profit optimization algorithm; According to the gross profit information and the risk value, obtaining the optimal shipment allocation quantity at each stage through an adaptive allocation optimization algorithm; Based on the gross profit accumulation at each stage, obtaining the global gross profit optimization information through a multi-dimensional progressive gross profit accumulation optimization algorithm; Based on the global gross profit optimization information, forming the feedback adjustment information of the supply chain system through an overall feedback closed-loop optimization algorithm.

2. The supply chain management method based on intelligent manufacturing according to claim 1, characterized in that The hierarchical weighted dynamic risk prediction algorithm includes: Generate a multi-dimensional risk factor vector V based on the cost information, operation and maintenance information, and inventory information at each stage i ; Use a weighted matrix W i to perform weighted processing on the risk factor vector to obtain a risk value R i ; The calculation formula for the risk value R i is as follows: Among them, R i is the risk value at the current stage, W i is the risk prediction weighted matrix, V i is the multi-dimensional risk factor vector within the stage, Q i is the risk threshold vector, α k is the influence weight of the risk factor, β k is the risk adjustment coefficient, M i is the actual gross profit of the stage, K i is the expected value of the gross profit prediction.

3. The supply chain management method based on intelligent manufacturing according to claim 2, characterized in that The superimposed non-linear gross profit optimization algorithm includes: Based on the risk value R i and the cost information and operation and maintenance information of the product, determine the gross profit G at the current stage i ; the gross profit G i is calculated by the formula: Among them, G i is the gross profit of the stage, S i is the sales volume of the stage, C i is the cost of the stage, M i is the operation and maintenance cost of the stage, λ is the non-linear adjustment factor, R i is the risk value output of the previous layer, and θ is the risk balance point.

4. The supply chain management method based on intelligent manufacturing according to claim 3, wherein The adaptive allocation optimization algorithm includes: According to the gross profit G i and the risk value R i , the shipment allocation quantity D is obtained through segmented adjustment i+1 ; The calculation formula for the shipment allocation quantity is as follows: Among them, D i+1 is the allocation amount in the (i + 1)-th stage, and D i is the basic allocation amount in the i-th stage, γ i is the risk allocation balance coefficient, δ is the risk threshold, and ζ is the adaptive allocation adjustment coefficient.

5. The supply chain management method based on intelligent manufacturing according to claim 4, wherein The multi-dimensional progressive gross profit accumulation optimization algorithm includes: Based on the gross profit information at each stage, combined with historical gross profit, perform progressive cumulative optimization of gross profit to obtain the global gross profit G total ; The global gross profit G total The calculation formula is as follows: Among them, G total is the cumulative gross profit, G i is the gross profit in the i-th stage, η i is the cumulative optimization factor of the stage, ξ is the gross profit growth sensitivity coefficient, G i-1 is the gross profit of the previous stage.

6. The supply chain management method based on intelligent manufacturing according to claim 5, wherein The overall feedback closed-loop optimization algorithm includes: Based on the global gross profit, combining the risk value and the allocation information, a feedback optimization quantity F is formed i ; The feedback optimization quantity F i has the following calculation formula: Among them, F i is the stage feedback optimization amount, G total is the cumulative gross profit, R j is the risk value at the j-th stage, ω j is the feedback adjustment coefficient.

7. A supply chain management system based on intelligent manufacturing for performing the method according to any one of claims 1-6, characterized in that, Including: An information acquisition module for obtaining the cost information, operation and maintenance information, and inventory information at each stage; A risk prediction module for obtaining the risk value at each stage based on the hierarchical weighted dynamic risk prediction algorithm; A gross profit calculation module for obtaining the gross profit information at each stage based on the superimposed non-linear gross profit optimization algorithm; An allocation optimization module for obtaining the shipment allocation quantity at each stage based on the adaptive allocation optimization algorithm; A gross profit accumulation module for obtaining the global gross profit information based on the multi-dimensional progressive gross profit accumulation optimization algorithm; A closed-loop feedback module for obtaining the feedback optimization information based on the overall feedback closed-loop optimization algorithm and updating it to the initial stage of the system.

8. The supply chain management system based on intelligent manufacturing according to claim 7, wherein The risk prediction module obtains the risk value at each stage based on the information obtained by the information acquisition module.

9. The supply chain management system based on intelligent manufacturing according to claim 8, wherein The gross profit calculation module obtains the gross profit information at each stage based on the risk value obtained by the risk prediction module.

10. The supply chain management system based on intelligent manufacturing according to claim 9, wherein The closed-loop feedback module generates the feedback optimization information based on the global gross profit information obtained by the gross profit accumulation module and updates it to the information acquisition module.

Images