Manufacturing industry supply chain optimization method, system and equipment based on large model and medium

Abstract

The invention discloses a manufacturing industry supply chain optimization method, system and equipment based on a large model, and a medium, belongs to the technical field of large model and supply chain management, and aims to solve the technical problem of how to monitor, predict and optimize a supply chain in real time, improve the efficiency of the supply chain and improve the efficiency of the supply chain. According to the technical scheme, the method comprises the following steps: data acquisition and preprocessing: comprehensively acquiring manufacturing industry supply chain data, and carrying out cleaning, duplicate removal, missing value processing, abnormal value detection and data normalization processing on the acquired manufacturing industry supply chain data to obtain preprocessed manufacturing industry supply chain data; building a large supply chain model: building the large supply chain model by utilizing the preprocessed manufacturing industry supply chain data, and training, optimizing and deploying the supply chain model; real-time monitoring and prediction: monitoring the inventory level and each index of the generation progress in the manufacturing industry supply chain in real time through the trained large model; optimizing the decision; and updating and iterating the model.

Description

Technical Field

[0001] The present invention relates to the technical field of large models and supply chain management, and in particular to a large model-based manufacturing supply chain optimization method, system, equipment and medium. Background Art

[0002] As the core link of manufacturing industry, the efficiency and response speed of manufacturing supply chain management directly affect the competitiveness and profitability of enterprises. How to optimize the supply chain and improve its efficiency and response speed has become an important issue facing the current manufacturing industry.

[0003] Traditional supply chain management methods mainly rely on manual experience and intuition for forecasting management, which has great limitations and often leads to problems such as information asymmetry, delayed decision-making, and waste of resources. On the one hand, traditional forecasting models are often trained based on small sample data, which makes it difficult to cope with complex and changing supply chain environments. On the other hand, traditional optimization methods can only consider local optimization goals and are difficult to achieve overall optimization.

[0004] Big model technology has achieved remarkable results in image recognition, natural language processing and other fields, providing new possibilities for manufacturing supply chain optimization. Supply chain optimization methods based on big models are gradually gaining attention.

[0005] Therefore, how to monitor, predict and optimize the supply chain in real time and improve the efficiency of the supply chain is a technical problem that needs to be solved urgently. Summary of the invention

[0006] The technical task of the present invention is to provide a manufacturing supply chain optimization method, system, equipment and medium based on a large model to solve the problem of how to monitor, predict and optimize the supply chain in real time and improve the efficiency of the supply chain.

[0007] The technical task of the present invention is achieved in the following way: a manufacturing supply chain optimization method based on a large model, the method is as follows:

[0008] Data collection and preprocessing: Comprehensively collect manufacturing supply chain data, clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data;

[0009] Build a large supply chain model: Use pre-processed manufacturing supply chain data to build a large supply chain model, and train, optimize and deploy the supply chain model;

[0010] Real-time monitoring and prediction: Use the trained big model to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time; use the supply chain big model to predict demand and inventory within a set time in the future, and provide accurate information support for corporate decision-making;

[0011] Optimization decision: Make corresponding optimization decisions based on the prediction results of the supply chain big model;

[0012] Model update and iteration: Based on data changes in the manufacturing supply chain, the supply chain model is updated regularly to adapt to data changes and business development in the manufacturing supply chain. At the same time, based on actual operating conditions, decisions are continuously adjusted and optimized to achieve better supply chain management results.

[0013] Preferably, the manufacturing supply chain data includes the following:

[0014] Sales data: collect information on product sales quantity, price, regional distribution and seasonal trends to understand market demand and consumer behavior;

[0015] Inventory data: record inventory levels, locations, expiration dates, and turnover rates for effective inventory management and forecasting;

[0016] Production data: monitor the operation status, production efficiency and product quality of the production line, optimize the production process and increase output;

[0017] Logistics data: Track the time, cost, route and delays of cargo transportation, optimize logistics networks and reduce transportation costs;

[0018] Supplier data: record suppliers’ delivery time, quality, price and creditworthiness, evaluate suppliers’ performance and select suitable suppliers;

[0019] Customer data: collect customer information, purchase history, preferences and feedback to personalize services and improve customer satisfaction;

[0020] Financial data: monitor costs, revenues, profits and cash flows, and conduct financial analysis and risk management;

[0021] Market data: collect information on competitors’ products, prices and promotions to develop effective marketing strategies;

[0022] Environmental data: Monitor environmental impacts in the supply chain, including carbon emissions and waste generation, to achieve sustainable development and a green supply chain.

[0023] Preferably, the supply chain model is constructed as follows:

[0024] Design model architecture: Design a suitable neural network model architecture based on the characteristics of the supply chain and optimization goals;

[0025] Select a training strategy: Select batch normalization, regularization, or dropout training strategies to improve the generalization ability of the model and avoid overfitting. When selecting a training strategy, also consider the hyperparameter settings of learning rate, batch size, and number of iterations.

[0026] Model training: Use optimization algorithms and loss functions to train the large supply chain model. During the training process, the convergence speed and performance of the large supply chain model are optimized by adjusting the learning rate and batch size.

[0027] Model tuning: Based on the results of performance evaluation, the supply chain model is tuned to improve its performance and accuracy through iterative training and tuning. Tuning the supply chain model includes adjusting the model architecture, training strategy, and hyperparameters.

[0028] Model deployment: Deploy the trained large supply chain model to the actual production environment, and monitor and update it in real time; in actual applications, consider the real-time and scalability of the model to ensure that it can efficiently process large amounts of data.

[0029] More preferably, the neural network model architecture includes a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), or a recurrent neural network (RNN) variant; wherein the recurrent neural network (RNN) includes a long short-term memory network (LSTM) or a gated recurrent unit (GRU).

[0030] Preferably, the optimization algorithm comprises stochastic gradient descent (SGD) or Adam;

[0031] Loss functions include mean squared error (MSE) or cross entropy loss.

[0032] As a preferred option, optimization decisions include adjusting production plans, optimizing inventory strategies and supplier selection; the goal of optimization decisions is to improve supply chain efficiency, reduce costs, improve service levels, and achieve optimized supply chain operations.

[0033] A manufacturing supply chain optimization system based on a large model, the system comprising:

[0034] The data collection and preprocessing module is used to comprehensively collect manufacturing supply chain data, and to clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data;

[0035] The supply chain big model building module is used to build a supply chain big model using pre-processed manufacturing supply chain data, and to train, optimize and deploy the supply chain model;

[0036] The real-time monitoring and prediction module is used to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time through the trained big model. At the same time, the supply chain big model is used to predict the demand and inventory situation within a set time in the future, providing accurate information support for enterprise decision-making;

[0037] The optimization decision module is used to make corresponding optimization decisions based on the prediction results of the supply chain big model;

[0038] The model update and iteration module is used to regularly update the supply chain model according to data changes in the manufacturing supply chain to adapt to data changes and business development in the manufacturing supply chain; at the same time, it continuously adjusts and optimizes decisions based on actual operating conditions to achieve better supply chain management results.

[0039] Preferably, the supply chain big model building module includes:

[0040] The model architecture design submodule is used to design a suitable neural network model architecture according to the characteristics of the supply chain and the optimization objectives;

[0041] The training strategy selection submodule is used to select batch normalization, regularization or dropout training strategies to improve the generalization ability of the model and avoid overfitting. When selecting a training strategy, the hyperparameter settings of learning rate, batch size and number of iterations are also considered.

[0042] The model training submodule is used to train the large supply chain model using the optimization algorithm and loss function. During the training process, the convergence speed and performance of the large supply chain model are optimized by adjusting the learning rate and batch size.

[0043] The model tuning submodule is used to tune the supply chain model based on the results of performance evaluation. Through iterative training and tuning, the performance and accuracy of the supply chain model can be improved. The content of supply chain model tuning includes adjusting the model architecture, training strategy and hyperparameters.

[0044] The model deployment submodule is used to deploy the trained supply chain model to the actual production environment and conduct real-time monitoring and updates. In actual applications, the real-time and scalability of the model are considered to ensure that large amounts of data can be processed efficiently.

[0045] Among them, the supply chain big model construction module uses machine learning or deep learning methods to build a predictive model.

[0046] An electronic device comprising: a memory and at least one processor;

[0047] Wherein, the memory stores computer-executable instructions;

[0048] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the manufacturing supply chain optimization method based on a large model as described above.

[0049] A computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the above-mentioned large-model-based manufacturing supply chain optimization method is implemented.

[0050] The manufacturing supply chain optimization method, system, device and medium based on a large model of the present invention have the following advantages:

[0051] (1) The present invention realizes real-time monitoring, prediction and optimization of the supply chain by constructing a large model, providing manufacturing enterprises with a more efficient and intelligent supply chain management solution, which helps to enhance the competitiveness and profitability of enterprises, thereby improving the efficiency of the supply chain, reducing costs, and improving service levels, and has good practical value and application prospects;

[0052] (ii) The present invention improves the efficiency of the supply chain: through real-time monitoring and forecasting, enterprises can better grasp market demand and adjust production plans in a timely manner, thereby improving production efficiency;

[0053] (III) The present invention reduces costs: by optimizing inventory strategies, enterprises can reduce inventory backlogs and reduce inventory costs;

[0054] (iv) The present invention improves service levels: through real-time monitoring and prediction, enterprises can better meet customer needs and improve customer satisfaction;

[0055] (V) The present invention solves the problem of information asymmetry: by collecting and analyzing a large amount of supply chain data, establishing a forecasting model, and realizing real-time sharing and transmission of information, the problem of information asymmetry is solved;

[0056] (VI) The present invention achieves global optimization: by building a large model and considering the global optimization goal, the overall optimization is achieved and the comprehensive performance of the supply chain is improved;

[0057] (VII) The present invention improves the competitiveness of enterprises: by optimizing the supply chain, improving efficiency, reducing costs, and improving service levels, the competitiveness and profitability of enterprises are improved;

[0058] (VIII) The present invention can respond to market changes: through real-time monitoring and prediction, enterprises can quickly respond to market changes, flexibly adjust supply chain strategies, and maintain the competitiveness of enterprises;

[0059] (IX) The invention promotes industry development: The application of this method will promote the development of manufacturing supply chain management in a more intelligent and efficient direction, and promote the progress and innovation of the entire industry;

[0060] (10) The present invention adopts innovative large model technology, combines multiple methods such as deep learning, big data analysis and operational optimization, and has the advantages of high efficiency, accuracy and scalability; among them, large model technology refers to a method of mining and analyzing data using a large-scale neural network model, which combines multiple technologies such as deep learning, big data analysis and operational optimization, and has the advantages of high efficiency, accuracy and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The present invention is further described below in conjunction with the accompanying drawings.

[0062] Attached Figure 1 A flowchart of the manufacturing supply chain optimization method based on a large model;

[0063] Attached Figure 2 A flowchart for feedback upgrade iteration. DETAILED DESCRIPTION

[0064] The following detailed description is made of the manufacturing supply chain optimization method, system, device and medium based on a large model of the present invention with reference to the drawings and specific embodiments of the specification.

[0065] Embodiment 1:

[0066] As attached Figure 1 As shown, this embodiment provides a manufacturing supply chain optimization method based on a large model, and the method is specifically as follows:

[0067] S1. Data collection and preprocessing: Comprehensively collect manufacturing supply chain data, and clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data;

[0068] S2. Build a large supply chain model: Use the pre-processed manufacturing supply chain data to build a large supply chain model, and train, optimize and deploy the supply chain model;

[0069] S3. Real-time monitoring and prediction: Use the trained big model to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time; use the supply chain big model to predict demand and inventory within a set time in the future, and provide accurate information support for corporate decision-making;

[0070] S4. Optimization decision: Make corresponding optimization decisions based on the prediction results of the supply chain big model;

[0071] S5. Model update and iteration: as shown in the attached Figure 2As shown, according to the data changes in the manufacturing supply chain, the supply chain big model is updated regularly to adapt to the data changes in the manufacturing supply chain and business development; at the same time, according to the actual operation situation, the decision is continuously adjusted and optimized to achieve better supply chain management results.

[0072] The manufacturing supply chain data in step S1 of this embodiment includes the following:

[0073] Sales data: collect information on product sales quantity, price, regional distribution and seasonal trends to understand market demand and consumer behavior;

[0074] Inventory data: record inventory levels, locations, expiration dates, and turnover rates for effective inventory management and forecasting;

[0075] Production data: monitor the operation status, production efficiency and product quality of the production line, optimize the production process and increase output;

[0076] Logistics data: Track the time, cost, route and delays of cargo transportation, optimize logistics networks and reduce transportation costs;

[0077] Supplier data: record suppliers’ delivery time, quality, price and creditworthiness, evaluate suppliers’ performance and select suitable suppliers;

[0078] Customer data: collect customer information, purchase history, preferences and feedback to personalize services and improve customer satisfaction;

[0079] Financial data: monitor costs, revenues, profits and cash flows, and conduct financial analysis and risk management;

[0080] Market data: collect information on competitors’ products, prices and promotions to develop effective marketing strategies;

[0081] Environmental data: Monitor environmental impacts in the supply chain, including carbon emissions and waste generation, to achieve sustainable development and a green supply chain.

[0082] The construction of the supply chain model in step S2 of this embodiment is as follows:

[0083] S201. Design model architecture: Design a suitable neural network model architecture based on the characteristics of the supply chain and optimization goals;

[0084] S202, select a training strategy: select batch normalization, regularization or dropout training strategy to improve the generalization ability of the model and avoid overfitting; when selecting a training strategy, the hyperparameter settings of the learning rate, batch size and number of iterations are also considered;

[0085] S203, model training: Use optimization algorithms and loss functions to train the supply chain model. During the training process, the convergence speed and performance of the supply chain model are optimized by adjusting the learning rate and batch size.

[0086] S204, Model Tuning: Tune the supply chain model based on the results of performance evaluation, and improve the performance and accuracy of the supply chain model through iterative training and tuning. The content of supply chain model tuning includes adjusting the model architecture, training strategy and hyperparameters.

[0087] S205, Model deployment: Deploy the trained supply chain large model to the actual production environment, and monitor and update it in real time; in actual applications, consider the real-time and scalability of the model to ensure that it can efficiently process large amounts of data.

[0088] The neural network model architecture in step S201 of this embodiment includes a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), or a recurrent neural network (RNN) variant; wherein the recurrent neural network (RNN) includes a long short-term memory network (LSTM) or a gated recurrent unit (GRU).

[0089] The optimization algorithm in step S203 of this embodiment includes stochastic gradient descent (SGD) or Adam.

[0090] The loss function in step S203 of this embodiment includes mean square error (MSE) or cross entropy loss.

[0091] The optimization decision in step S4 of this embodiment includes adjusting the production plan, optimizing the inventory strategy and supplier selection; the goal of the optimization decision is to improve the efficiency of the supply chain, reduce costs, improve service levels, and achieve optimized operation of the supply chain.

[0092] Embodiment 2:

[0093] This embodiment provides a manufacturing supply chain optimization system based on a large model, the system comprising:

[0094] The data collection and preprocessing module is used to comprehensively collect manufacturing supply chain data, and to clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data;

[0095] The supply chain big model building module is used to build a supply chain big model using pre-processed manufacturing supply chain data, and to train, optimize and deploy the supply chain model;

[0096] The real-time monitoring and prediction module is used to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time through the trained big model. At the same time, the supply chain big model is used to predict the demand and inventory situation within a set time in the future, providing accurate information support for enterprise decision-making;

[0097] The optimization decision module is used to make corresponding optimization decisions based on the prediction results of the supply chain big model;

[0098] The model update and iteration module is used to regularly update the supply chain model according to data changes in the manufacturing supply chain to adapt to data changes and business development in the manufacturing supply chain; at the same time, it continuously adjusts and optimizes decisions based on actual operating conditions to achieve better supply chain management results.

[0099] The supply chain big model building module in this embodiment includes:

[0100] The model architecture design submodule is used to design a suitable neural network model architecture according to the characteristics of the supply chain and the optimization objectives;

[0101] The training strategy selection submodule is used to select batch normalization, regularization or dropout training strategies to improve the generalization ability of the model and avoid overfitting. When selecting a training strategy, the hyperparameter settings of learning rate, batch size and number of iterations are also considered.

[0102] The model training submodule is used to train the large supply chain model using the optimization algorithm and loss function. During the training process, the convergence speed and performance of the large supply chain model are optimized by adjusting the learning rate and batch size.

[0103] The model tuning submodule is used to tune the supply chain model based on the results of performance evaluation. Through iterative training and tuning, the performance and accuracy of the supply chain model can be improved. The content of supply chain model tuning includes adjusting the model architecture, training strategy and hyperparameters.

[0104] The model deployment submodule is used to deploy the trained supply chain model to the actual production environment and conduct real-time monitoring and updates. In actual applications, the real-time and scalability of the model are considered to ensure that large amounts of data can be processed efficiently.

[0105] Among them, the supply chain big model construction module uses machine learning or deep learning methods to build a predictive model.

[0106] Embodiment 3:

[0107] This embodiment also provides an electronic device, including: a memory and at least one processor;

[0108] Wherein, the memory stores computer-executable instructions;

[0109] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the large model-based manufacturing supply chain optimization method described in any one of the present inventions.

[0110] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or any conventional processor, etc.

[0111] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory can also include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash memory card, at least one disk storage period, a flash memory device, or other volatile solid-state storage devices.

[0112] Embodiment 4:

[0113] This embodiment also provides a computer-readable storage medium, which stores a plurality of instructions, which are loaded by a processor, so that the processor executes the large model-based manufacturing supply chain optimization method in any embodiment of the present invention. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code that implements the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0114] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0115] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0116] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0117] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A manufacturing supply chain optimization method based on a large model, characterized in that: The method is as follows: Data collection and preprocessing: Comprehensively collect manufacturing supply chain data, clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data; Build a large supply chain model: Use pre-processed manufacturing supply chain data to build a large supply chain model, and train, optimize and deploy the supply chain model; Real-time monitoring and prediction: Use the trained big model to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time; use the supply chain big model to predict demand and inventory within a set time in the future, and provide accurate information support for corporate decision-making; Optimization decision: Make corresponding optimization decisions based on the prediction results of the supply chain big model; Model update and iteration: Based on data changes in the manufacturing supply chain, the supply chain model is updated regularly to adapt to data changes and business development in the manufacturing supply chain. At the same time, based on actual operating conditions, decisions are continuously adjusted and optimized to achieve better supply chain management results.

2. The manufacturing supply chain optimization method based on a large model according to claim 1 is characterized in that: Manufacturing supply chain data includes the following: Sales data: collect information on product sales quantity, price, regional distribution and seasonal trends to understand market demand and consumer behavior; Inventory data: record inventory levels, locations, expiration dates, and turnover rates for effective inventory management and forecasting; Production data: monitor the operation status, production efficiency and product quality of the production line, optimize the production process and increase output; Logistics data: Track the time, cost, route and delays of cargo transportation, optimize logistics networks and reduce transportation costs; Supplier data: record suppliers’ delivery time, quality, price and creditworthiness, evaluate suppliers’ performance and select suitable suppliers; Customer data: collect customer information, purchase history, preferences and feedback to personalize services and improve customer satisfaction; Financial data: monitor costs, revenues, profits and cash flows, and conduct financial analysis and risk management; Market data: collect information on competitors’ products, prices and promotions to develop effective marketing strategies; Environmental data: Monitor environmental impacts in the supply chain, including carbon emissions and waste generation, to achieve sustainable development and a green supply chain.

3. The manufacturing supply chain optimization method based on a large model according to claim 1 or 2, characterized in that: The details of building a supply chain model are as follows: Design model architecture: Design a suitable neural network model architecture based on the characteristics of the supply chain and optimization goals; Select a training strategy: Select batch normalization, regularization, or dropout training strategies to improve the generalization ability of the model and avoid overfitting. When selecting a training strategy, also consider the hyperparameter settings of learning rate, batch size, and number of iterations. Model training: Use optimization algorithms and loss functions to train the large supply chain model. During the training process, the convergence speed and performance of the large supply chain model are optimized by adjusting the learning rate and batch size. Model tuning: Based on the results of performance evaluation, the supply chain model is tuned to improve its performance and accuracy through iterative training and tuning. Tuning the supply chain model includes adjusting the model architecture, training strategy, and hyperparameters. Model deployment: Deploy the trained supply chain model to the actual production environment and perform real-time monitoring and updates.

4. The manufacturing supply chain optimization method based on a large model according to claim 3 is characterized in that: The neural network model architecture includes a multi-layer perceptron, a convolutional neural network, a recurrent neural network, or a variant of a recurrent neural network; wherein the recurrent neural network includes a long short-term memory network or a gated recurrent unit.

5. The manufacturing supply chain optimization method based on a large model according to claim 3 is characterized in that: Optimization algorithms include stochastic gradient descent or Adam; Loss functions include mean squared error (MSE) or cross entropy loss.

6. The manufacturing supply chain optimization method based on a large model according to claim 1 is characterized in that: Optimization decisions include adjusting production plans, optimizing inventory strategies and supplier selection.

7. A manufacturing supply chain optimization system based on a large model, characterized in that: The system includes: The data collection and preprocessing module is used to comprehensively collect manufacturing supply chain data, and to clean, remove duplicates, process missing values, detect outliers, and normalize the collected manufacturing supply chain data to obtain preprocessed manufacturing supply chain data; The supply chain big model building module is used to build a supply chain big model using pre-processed manufacturing supply chain data, and to train, optimize and deploy the supply chain model; The real-time monitoring and prediction module is used to monitor the inventory level and various indicators of the production progress in the manufacturing supply chain in real time through the trained big model. At the same time, the supply chain big model is used to predict the demand and inventory situation within a set time in the future, providing accurate information support for enterprise decision-making; The optimization decision module is used to make corresponding optimization decisions based on the prediction results of the supply chain big model; The model update and iteration module is used to regularly update the supply chain model according to data changes in the manufacturing supply chain to adapt to data changes and business development in the manufacturing supply chain; at the same time, it continuously adjusts and optimizes decisions based on actual operating conditions to achieve better supply chain management results.

8. The manufacturing supply chain optimization system based on a large model according to claim 7 is characterized in that: The supply chain big model building modules include: The model architecture design submodule is used to design a suitable neural network model architecture according to the characteristics of the supply chain and the optimization objectives; The training strategy selection submodule is used to select batch normalization, regularization or dropout training strategies to improve the generalization ability of the model and avoid overfitting. When selecting a training strategy, the hyperparameter settings of learning rate, batch size and number of iterations are also considered. The model training submodule is used to train the large supply chain model using the optimization algorithm and loss function. During the training process, the convergence speed and performance of the large supply chain model are optimized by adjusting the learning rate and batch size. The model tuning submodule is used to tune the supply chain model based on the results of performance evaluation. Through iterative training and tuning, the performance and accuracy of the supply chain model can be improved. The content of supply chain model tuning includes adjusting the model architecture, training strategy and hyperparameters. The model deployment submodule is used to deploy the trained supply chain large model to the actual production environment and perform real-time monitoring and updates; Among them, the supply chain big model construction module uses machine learning or deep learning methods to build a predictive model.

9. An electronic device, characterized in that: include: memory and at least one processor; Wherein, the memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the large model-based manufacturing supply chain optimization method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the processor executes the computer-executable instructions, the large-model-based manufacturing supply chain optimization method as described in any one of claims 1 to 6 is implemented.

Images