Management system and method for digital intelligent factory

Abstract

The invention relates to the technical field of management of intelligent factories, in particular to a management system and method of a digital intelligent factory. The method comprises the following steps: a production management unit predicts demands, plans production and allocates resources by using an SVM machine learning algorithm model; the quality management unit ensures that the product quality meets the standard and customer requirements through real-time monitoring and data analysis; the equipment management unit monitors the equipment state in real time by using the Internet of Things integration technology; the inventory management unit monitors the inventory level in real time; the data analysis and decision support unit adopts a stream processing technology, analyzes data from all units in real time, and provides decision support. According to the system, an Internet of Things integration method is adopted, automation equipment and a robot technology are combined with a digital twinning technology, the Internet of Things, automation, the robot technology and digital twinning are integrated, comprehensive digitization of the production process is achieved, and the production efficiency is improved. And technologies of different sources and types are effectively integrated into an existing production environment to realize automatic management.

Description

Technical Field

[0001] The present invention relates to the technical field of smart factory management, and in particular to a management system for a digital smart factory. Background Art

[0002] With the development of science and technology and industrial progress, digital smart factories have become a key trend in modern manufacturing. By applying advanced information technology and automation systems, digital smart factories achieve digitization, automation, and intelligence in production processes, improving production efficiency and quality while reducing costs and resource consumption. Building digital smart factories is an inevitable trend in the development of modern manufacturing and a key means of improving production efficiency and quality. The construction and operation of digital smart factories can be achieved through measures such as building stable and reliable infrastructure, digitizing and automating production processes, applying intelligent manufacturing and collaborative robotics technologies, and optimizing management and operations. However, building digital smart factories also presents challenges. These challenges require comprehensive consideration of various factors, effectively integrating diverse technologies into the existing production environment for automated management, and overcoming challenges in technical compatibility and data exchange. Therefore, a management system and method for digital smart factories are designed. Summary of the Invention

[0003] The purpose of the present invention is to provide a management system and method for a digital smart factory to solve the problem of effectively integrating technologies of different sources and types into the existing production environment to achieve automated management proposed in the above background technology, which requires overcoming technical compatibility and data exchange problems.

[0004] To achieve the above objectives, on the one hand, the present invention aims to provide a management system for a digital smart factory, comprising: a production management unit, which uses an SVM machine learning algorithm model to predict demand, plan production, and allocate resources;

[0005] A quality management unit, which ensures that product quality meets standards and customer requirements through real-time monitoring and data analysis, and promptly identifies and resolves quality issues;

[0006] An equipment management unit, which uses IoT integration technology to monitor equipment status in real time, performs maintenance, upkeep, and troubleshooting tasks based on feedback from the quality management unit, and provides equipment status information to the production management unit;

[0007] An inventory management unit that monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis;

[0008] The logistics management unit plans procurement and transportation plans based on the inventory levels and forecast data of the inventory management unit, and coordinates with the production management unit to ensure that finished products are delivered to customers on time;

[0009] The data analysis and decision support unit uses stream processing technology to analyze data from all units in real time and provide decision support.

[0010] As a further improvement of this technical solution, the production management unit adopts an SVM machine learning algorithm model to dynamically adjust the production plan based on historical data and real-time conditions;

[0011] Among them, historical data includes production records, quality control data, equipment operation data, inventory change records, supply chain information, financial data, and market demand data;

[0012] Real-time data includes production progress data, equipment status data, inventory level data, and supply chain data.

[0013] Among them, the SVM machine learning algorithm model is a supervised learning model widely used in classification and regression analysis. The specific steps are:

[0014] S3.1. Collect data, convert categorical data into numerical data, and perform feature scaling.

[0015] S3.2. Identify the key features that affect production planning and conduct feature selection;

[0016] S3.3. Use the historical data set to split the training set and test set, select the kernel function, and use the training set data to train the SVM model;

[0017] S3.4. Use the test set data to evaluate the model performance and adjust the SVM parameters based on the evaluation results.

[0018] S3.5. Collect real-time data as input, use the trained SVM model to make predictions, and obtain adjustment suggestions for the production plan.

[0019] As a further improvement of this technical solution, the quality management unit ensures that product quality meets standards and customer requirements through real-time monitoring and data analysis, promptly discovers and resolves quality issues, and ensures continuous improvement and optimization of product quality, thereby improving customer satisfaction and the company's market competitiveness;

[0020] As a further improvement of this technical solution, the device management unit adopts Internet of Things integration to connect physical devices, sensors, actuators and smart devices through the Internet to manage the devices;

[0021] IoT integration is a complex process that involves connecting physical devices, sensors, actuators, and smart devices via the internet to enable data exchange and communication. Key IoT integration functions include enabling diverse applications to easily communicate and exchange data, often involving the integration and use of APIs to ensure interoperability. IoT platforms typically offer cloud-based APIs, allowing servers to call APIs via cloud-based SDKs to send instructions to devices for remote control and monitoring. IoT integration requires not only precise technical integration but also comprehensive considerations of data processing, security, scalability, and interoperability. Effective IoT integration enables intelligent connectivity between devices, providing a solid foundation for building digital smart factories.

[0022] As a further improvement of this technical solution, the inventory management unit includes a material demand module and a warehouse control module;

[0023] Among them, the material demand module uses exponential smoothing method to predict future demand based on past data;

[0024] The warehouse control module uses automated equipment and robotics technology, combined with digital twin technology, to achieve automated warehouse management and human-computer interaction.

[0025] Furthermore, exponential smoothing is a statistical method used for time series forecasting. It makes predictions by assigning different weights to past observations. It is simple and adaptable. When processing time series data, it can effectively capture the latest trends of the data. Especially when the data volume is small or the data fluctuations are not large, the exponential smoothing method can provide more accurate forecast results. Its specific formula is:

[0026] S t =a*y t +(1-a)*S t-1 ;

[0027] Among them, S t is the predicted value for period t, y t is the actual value of period t, S t-1 is the predicted value of the t-1 period, a is a smoothing constant, and its value range is [0,1].

[0028] Furthermore, digital twins are a technology that creates virtual models of physical objects, systems, or processes that reflect the status of their corresponding entities in real time. In this way, various scenarios can be tested, analyzed, and optimized without affecting actual operations. At the same time, by collecting data and using advanced simulation software, digital twins can predict the performance and failure of equipment, thereby optimizing maintenance plans.

[0029] Combining automation equipment and robotics with digital twins can create a highly synchronized and interactive production environment. Real-time data is transmitted from robots and automation equipment on the production line to the digital twin model, allowing real-time monitoring and analysis. Utilizing the data analysis and simulation capabilities of the digital twin model, the production process can be optimized, potential failures can be predicted and prevented, thereby improving equipment availability and production line reliability.

[0030] Among them, the specific steps to be achieved by using automated equipment and robotics technology combined with digital twin technology are:

[0031] S6.1. Deploy automation equipment and robotics in warehouses;

[0032] S6.2. Collect equipment and warehouse information and build a digital twin model;

[0033] S6.3. Design human-computer interaction interface to control equipment and robots.

[0034] Furthermore, the formula involved in establishing the digital twin model is:

[0035] Use a linear regression model to predict the state or position of a device:

[0036] Y=β0+β1*X;

[0037] Where Y is the target variable to be predicted, X is the input variable, β0 is the intercept term, and β1 is the slope parameter;

[0038] Use the least squares method to estimate the model parameters:

[0039] L=∑(Y i -(β0+β1*X i ))^2;

[0040] Among them, L is the loss function, Y i is the actual value of the i-th sample, X i is the input value of the i-th sample.

[0041] As a further improvement of this technical solution, the logistics management unit includes a transportation management module, a supplier management module, and a customer order management module;

[0042] Among them, the transportation management module uses the Dijkstra algorithm to plan and optimize the material transportation process; the supplier management module uses the electronic procurement system to manage suppliers, evaluate and select suppliers, manage purchase orders and deliveries; the customer order management module uses the customer relationship management system to manage customer orders, receive and process customer orders, and track order status.

[0043] Furthermore, the Dijkstra algorithm is a classic graph search algorithm that can be used to solve the shortest path problem. In transportation management, the Dijkstra algorithm can help determine the lowest-cost or shortest-time transportation route from a starting point to a destination within a given road network. By integrating real-time traffic information, vehicle capacity, delivery deadlines, and cost factors, the transportation management module can optimize cargo distribution plans, reduce transportation costs, and improve transportation efficiency. An e-procurement system is an electronic platform that simplifies and automates the procurement process. Through the e-procurement system, companies can publish procurement requirements, receive supplier quotations, conduct online negotiations, place purchase orders, and track order status. Furthermore, the system collects supplier performance data, such as delivery on-time performance, quality compliance, and service levels, helping companies assess supplier reliability and performance, thereby making more informed supplier selection decisions. A customer relationship management system is an integrated software platform used to manage interactions between companies and their customers. Through a customer relationship management system, companies can conveniently record customer information, order details, historical transactions, and service requests.

[0044] Furthermore, the specific formula of Dijkstra's algorithm is:

[0045] For each vertex (v)∈(L(u)), if

[0046] D(v)>D(u)+w(u,v);

[0047] Update

[0048] D(v)=D(u)+w(u,v);

[0049] Where D(v) is the estimated distance of the shortest path from the starting point s to the vertex v, L(v) is the neighbor list of vertex v, and w(u,v) is the weight of the edge from vertex u to v.

[0050] As a further improvement of this technical solution, the data analysis and decision support unit uses stream processing technology to realize real-time analysis of real-time data streams and make decisions;

[0051] Stream processing is a technology that processes high-speed, continuous data streams. Compared with traditional batch processing, stream processing can provide near real-time data analysis, which is particularly important for application scenarios that need to respond quickly to market changes or operating conditions. In a stream processing architecture, data streams are captured in real time and delivered to the processing system, which includes event detection, aggregation, transformation, and related analysis algorithms.

[0052] In another aspect, the present invention provides a management method for a digital smart factory, based on any one of the digital smart factory management systems described above, comprising the following steps:

[0053] S9.1. The production management unit uses the SVM machine learning algorithm model to predict future product demand based on historical data and market trends, and plans production activities and resource allocation accordingly;

[0054] S9.2. The equipment management unit uses IoT technology to collect real-time data on the equipment's operating status and feeds this data back to the production management unit. The production management unit then adjusts the production plan in real time based on the equipment status report.

[0055] S9.3. The quality management unit monitors quality data during the production process in real time and cooperates with the production management unit;

[0056] S9.4. The equipment management unit shall perform necessary maintenance, servicing, and troubleshooting tasks based on the equipment status and feedback from the quality management unit;

[0057] S9.5. The inventory management unit monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis results;

[0058] S9.6. The logistics management unit plans procurement and transportation plans based on the inventory levels and forecast data of the inventory management unit;

[0059] S9.7. The data analysis and decision support unit uses stream processing technology to analyze data from all units in real time, providing management with a comprehensive data view and decision support.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] 1. This digital smart factory management system and method adopts an Internet of Things integration method, and combines automation equipment and robotics technology with digital twin technology. It integrates the Internet of Things, automation, robotics and digital twins, and realizes the comprehensive digitalization and intelligence of the production process, greatly improving the factory's operational efficiency, product quality and market response speed, while reducing operating costs and safety risks, and effectively integrating technologies from different sources and types into the existing production environment to realize automated management.

[0062] 2. This digital smart factory management system and method uses an SVM algorithm model to dynamically adjust production plans and uses exponential smoothing to predict product demand, achieving more accurate data analysis and decision support, improving production efficiency, reducing operating costs, and enhancing product quality and customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flowchart of the overall process of the present invention.

[0064] The meaning of each number in the figure is:

[0065] 1. Production management unit; 2. Quality management unit; 3. Equipment management unit; 4. Inventory management unit; 5. Logistics management unit; 6. Data analysis and decision support unit. DETAILED DESCRIPTION

[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0067] Example 1:

[0068] See also Figure 1 As shown, a management system for a digital smart factory is provided, including a production management unit 1, wherein the production management unit 1 uses an SVM machine learning algorithm model to predict demand, plan production, and allocate resources;

[0069] Quality management unit 2, which ensures that product quality meets standards and customer requirements through real-time monitoring and data analysis, and promptly identifies and resolves quality issues;

[0070] Equipment management unit 3, which uses IoT integration technology to monitor equipment status in real time, performs maintenance, upkeep, and troubleshooting tasks based on feedback from quality management unit 2, and provides equipment status information to production management unit 1;

[0071] Inventory management unit 4, which monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis;

[0072] Logistics management unit 5, which plans procurement and transportation plans based on the inventory levels and forecast data of inventory management unit 4, and coordinates with production management unit 1 to ensure that finished products are delivered to customers on time;

[0073] The data analysis and decision support unit 6 adopts stream processing technology to analyze data from all units in real time and provide decision support.

[0074] In this embodiment, the production management unit 1 uses an SVM machine learning algorithm model to dynamically adjust the production plan based on historical data and real-time conditions;

[0075] Among them, historical data includes production records, quality control data, equipment operation data, inventory change records, supply chain information, financial data, and market demand data;

[0076] Real-time data includes production progress data, equipment status data, inventory level data, and supply chain data.

[0077] In this embodiment, the SVM machine learning algorithm model is a supervised learning model widely used in classification and regression analysis. The specific steps are:

[0078] S3.1. Collect data, convert categorical data into numerical data, and perform feature scaling.

[0079] S3.2. Identify the key features that affect production planning and conduct feature selection;

[0080] S3.3. Use the historical data set to split the training set and test set, select the kernel function, and use the training set data to train the SVM model;

[0081] S3.4. Use the test set data to evaluate the model performance and adjust the SVM parameters based on the evaluation results.

[0082] S3.5. Collect real-time data as input, use the trained SVM model to make predictions, and obtain adjustment suggestions for the production plan.

[0083] The digital smart factory management system also includes a quality management unit 2, which ensures that product quality meets standards and customer requirements through real-time monitoring and data analysis, promptly identifies and resolves quality issues, and ensures continuous improvement and optimization of product quality, thereby enhancing customer satisfaction and the company's market competitiveness.

[0084] The management system of the digital smart factory also includes a device management unit 3, which uses the Internet of Things integration to connect physical devices, sensors, actuators and smart devices through the Internet to manage the devices;

[0085] IoT integration is a complex process that involves connecting physical devices, sensors, actuators, and smart devices via the internet to enable data exchange and communication. Key IoT integration functions include enabling diverse applications to easily communicate and exchange data, often involving the integration and use of APIs to ensure interoperability. IoT platforms typically offer cloud-based APIs, allowing servers to call APIs via cloud-based SDKs to send instructions to devices for remote control and monitoring. IoT integration requires not only precise technical integration but also comprehensive considerations of data processing, security, scalability, and interoperability. Effective IoT integration enables intelligent connectivity between devices, providing a solid foundation for building digital smart factories.

[0086] The management system of the digital smart factory also includes an inventory management unit 4, which includes a material demand module and a warehouse control module;

[0087] Among them, the material demand module uses exponential smoothing method to predict future demand based on past data;

[0088] The warehouse control module uses automated equipment and robotics technology, combined with digital twin technology, to achieve automated warehouse management and human-computer interaction.

[0089] Furthermore, exponential smoothing is a statistical method used for time series forecasting. It makes predictions by assigning different weights to past observations. It is simple and adaptable. When processing time series data, it can effectively capture the latest trends of the data. Especially when the data volume is small or the data fluctuations are not large, the exponential smoothing method can provide more accurate forecast results. Its specific formula is:

[0090] S t =a*y t +(1-a)*S t-1 ;

[0091] Among them, S t is the predicted value for period t, y t is the actual value of period t, S t-1 is the predicted value of the t-1 period, a is a smoothing constant, and its value range is [0,1].

[0092] Furthermore, digital twins are a technology that creates virtual models of physical objects, systems, or processes that reflect the status of their corresponding entities in real time. In this way, various scenarios can be tested, analyzed, and optimized without affecting actual operations. At the same time, by collecting data and using advanced simulation software, digital twins can predict the performance and failure of equipment, thereby optimizing maintenance plans.

[0093] Combining automation equipment and robotics with digital twins can create a highly synchronized and interactive production environment. Real-time data is transmitted from robots and automation equipment on the production line to the digital twin model, allowing real-time monitoring and analysis. Utilizing the data analysis and simulation capabilities of the digital twin model, the production process can be optimized, potential failures can be predicted and prevented, thereby improving equipment availability and production line reliability.

[0094] In this embodiment, the specific steps implemented using automated equipment and robotics combined with digital twin technology are as follows:

[0095] S6.1. Deploy automation equipment and robotics in warehouses;

[0096] S6.2. Collect equipment and warehouse information and build a digital twin model;

[0097] S6.3. Design human-computer interaction interface to control equipment and robots.

[0098] Furthermore, the formula involved in establishing the digital twin model is:

[0099] Use a linear regression model to predict the state or position of a device:

[0100] Y=β0+β1*X;

[0101] Where Y is the target variable to be predicted, X is the input variable, β0 is the intercept term, and β1 is the slope parameter;

[0102] Use the least squares method to estimate the model parameters:

[0103] L=∑(Y i -(β0+β1*X i ))^2;

[0104] Among them, L is the loss function, Y i is the actual value of the i-th sample, X i is the input value of the i-th sample.

[0105] The management system of the digital smart factory also includes a logistics management unit 5, which includes a transportation management module, a supplier management module, and a customer order management module;

[0106] Among them, the transportation management module uses the Dijkstra algorithm to plan and optimize the material transportation process; the supplier management module uses the electronic procurement system to manage suppliers, evaluate and select suppliers, manage purchase orders and deliveries; the customer order management module uses the customer relationship management system to manage customer orders, receive and process customer orders, and track order status.

[0107] In this embodiment, the Dijkstra algorithm is a classic graph search algorithm that can be used to solve the shortest path problem. In transportation management, the Dijkstra algorithm can help determine the lowest-cost or shortest-time transportation route from a starting point to a destination within a given road network. By integrating real-time traffic information, vehicle capacity, delivery deadlines, and cost factors, the transportation management module can optimize cargo distribution plans, reduce transportation costs, and improve transportation efficiency. The e-procurement system is an electronic platform that simplifies and automates the procurement process. Through the e-procurement system, companies can publish procurement requirements, receive supplier quotations, conduct online negotiations, place purchase orders, and track order status. Furthermore, the system collects supplier performance data, such as delivery on-time performance, quality compliance, and service levels, helping companies assess supplier reliability and performance, thereby making more informed supplier selection decisions. A customer relationship management system is an integrated software platform used to manage interactions between companies and their customers. Through the CRM system, companies can conveniently record customer information, order details, historical transactions, and service requests.

[0108] Furthermore, the specific formula of Dijkstra's algorithm is:

[0109] For each vertex (v)∈(L(u)), if

[0110] D(v)>D(u)+w(u,v);

[0111] Update

[0112] D(v)=D(u)+w(u,v);

[0113] Where D(v) is the estimated distance of the shortest path from the starting point s to the vertex v, L(v) is the neighbor list of vertex v, and w(u, v) is the weight of the edge from vertex u to v.

[0114] The management system of the digital smart factory also includes a data analysis and decision support unit 6, which uses stream processing technology to realize real-time analysis of real-time data streams and make decisions;

[0115] Stream processing is a technology that processes high-speed, continuous data streams. Compared with traditional batch processing, stream processing can provide near real-time data analysis, which is particularly important for application scenarios that need to respond quickly to market changes or operating conditions. In a stream processing architecture, data streams are captured in real time and delivered to the processing system, which includes event detection, aggregation, transformation, and related analysis algorithms.

[0116] Example 2:

[0117] The difference between Example 2 of the present invention and Example 1 is that this example introduces a static mechanical properties collection and analysis method used in a management system of a digital intelligent factory.

[0118] A management method for a digital smart factory, based on the above-mentioned management system for a digital smart factory, comprises the following steps:

[0119] S9.1. Production Management Unit 1 uses the SVM machine learning algorithm model to predict future product demand based on historical data and market trends, and plans production activities and resource allocation accordingly;

[0120] S9.2. Equipment management unit 3 uses IoT technology to collect real-time operating status data of equipment and feeds this data back to production management unit 1. Production management unit 1 adjusts the production plan in real time based on the equipment status report.

[0121] S9.3. Quality management unit 2 monitors quality data during the production process in real time and cooperates with production management unit 1;

[0122] S9.4. The equipment management unit 3 performs necessary maintenance, servicing, and troubleshooting tasks based on the equipment status and feedback from the quality management unit 2;

[0123] S9.5, Inventory Management Unit 4 monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis results;

[0124] S9.6. Logistics management unit 5 plans procurement and transportation plans based on the inventory level and forecast data of inventory management unit 4;

[0125] S9.7, Data Analysis and Decision Support Unit 6 uses stream processing technology to analyze data from all units in real time, providing management with a comprehensive data view and decision support.

[0126] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital smart factory management system, characterized by: include A production management unit (1), wherein the production management unit (1) uses an SVM machine learning algorithm model to predict demand, plan production, and allocate resources; A quality management unit (2), which ensures that product quality meets standards and customer requirements through real-time monitoring and data analysis, and promptly identifies and resolves quality issues; An equipment management unit (3), wherein the equipment management unit (3) uses Internet of Things integration technology to monitor equipment status in real time, performs maintenance, care, and troubleshooting tasks based on feedback from the quality management unit (2), and provides equipment status reports to the production management unit (1); An inventory management unit (4), which monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis; a logistics management unit (5), wherein the logistics management unit (5) plans procurement and transportation plans based on the inventory level and forecast data of the inventory management unit (4), and coordinates with the production management unit (1) to ensure that the finished products are delivered to customers on time; The data analysis and decision support unit (6) adopts stream processing technology to analyze data from all units in real time and provide decision support.

2. The digital smart factory management system according to claim 1 is characterized by: The production management unit (1) adopts an SVM machine learning algorithm model to dynamically adjust the production plan based on historical data and real-time data; Among them, historical data includes production records, quality control data, equipment operation data, inventory change records, supply chain information, financial data, and market demand data; Real-time data includes production progress data, equipment status data, inventory level data, and supply chain data.

3. The digital smart factory management system according to claim 2, characterized in that: The SVM machine learning algorithm model has the following specific steps: S3.

1. Collect historical data, convert categorical data into numerical data, and perform feature scaling. S3.

2. Identify the key features that affect production planning and conduct feature selection; S3.

3. Use the historical data set to split the training set and test set, select the kernel function, and use the training set data to train the SVM model; S3.

4. Use the test set data to evaluate the model performance and adjust the SVM parameters based on the evaluation results. S3.

5. Collect real-time data as input, use the trained SVM model to make predictions, and obtain adjustment suggestions for the production plan.

4. The digital smart factory management system according to claim 1, characterized in that: The device management unit (3) uses Internet of Things integration technology to connect physical devices, sensors, actuators and smart devices through the Internet.

5. The digital smart factory management system according to claim 1 is characterized in that: The inventory management unit (4) includes a material demand module and a warehouse control module; Among them, the material demand module uses exponential smoothing method to predict future demand based on past data; The warehouse control module uses automated equipment and robotics technology, combined with digital twin technology, to achieve automated warehouse management and human-computer interaction.

6. The digital smart factory management system according to claim 5, characterized in that: The specific steps for using automated equipment and robotics technology, combined with digital twin technology, to achieve warehouse automation management and human-machine interaction are as follows: S6.

1. Deploy automation equipment and robotics in warehouses; S6.

2. Collect equipment and warehouse information and build a digital twin model; S6.

3. Design human-computer interaction interface to control equipment and robots.

7. The digital smart factory management system according to claim 6, characterized in that: The expressions involved in establishing the digital twin model in S6.2 are: Use a linear regression model to predict the state or position of a device: Y=β0+β1*X; Where Y is the target variable to be predicted, X is the input variable, β0 is the intercept term, and β1 is the slope parameter; Use the least squares method to estimate the model parameters: L=∑(Y i -(β0+β1*X i ))^2; Among them, L is the loss function, Y i is the actual value of the i-th sample, X i is the input value of the i-th sample.

8. The digital smart factory management system according to claim 1, characterized in that: The logistics management unit (5) includes a transportation management module, a supplier management module, and a customer order management module; Among them, the transportation management module uses the Dijkstra algorithm to plan and optimize the material transportation process; The supplier management module uses the electronic procurement system to manage suppliers, evaluate and select suppliers, and manage purchase orders and deliveries; The customer order management module uses the customer relationship management system to manage customer orders, receive and process customer orders, and track order status.

9. The digital smart factory management system according to claim 8, characterized in that: The specific formula of Dijkstra's algorithm is: For each vertex (v)∈(L(u)), if D(v)>D(u)+w(u,v); Update D(v)=D(u)+w(u,v); Where D(v) is the estimated distance of the shortest path from the starting point s to the vertex v, L(v) is the neighbor list of vertex v, and w(u,v) is the weight of the edge from vertex u to v.

10. A digital smart factory management method, based on a digital smart factory management system according to any one of claims 1 to 9, characterized in that: The steps include: S9.

1. Production Management Unit (1) uses the SVM machine learning algorithm model to predict future product demand based on historical data and market trends, and plans production activities and resource allocation accordingly; S9.2, the equipment management unit (3) uses the Internet of Things technology to collect the operating status data of the equipment in real time, and feeds this data back to the production management unit (1). The production management unit (1) adjusts the production plan in real time according to the equipment status report; S9.3, the quality management unit (2) monitors the quality data of the production process in real time and cooperates with the production management unit (1); S9.4, Equipment Management Unit (3) performs necessary maintenance, servicing, and troubleshooting tasks based on the equipment status and feedback from the Quality Management Unit; S9.5, Inventory Management Unit (4) monitors inventory levels in real time and adjusts inventory strategies based on sales data and forecast analysis results; S9.6, the logistics management unit (5) plans the purchase and transportation plan based on the inventory level and forecast data of the inventory management unit (4); S9.

7. The Data Analysis and Decision Support Unit (6) uses stream processing technology to analyze data from all units in real time, providing management with a comprehensive data view and decision support.

Images