A cloud computing-based factory operation data analysis method and system

Abstract

The application discloses a kind of factory operation data analysis method and system based on cloud computing, by obtaining multiple production factory big data and based on the data division of classification model of decision tree, and form product production data, production operation data, equipment operation data, manpower and material resources allocation data, product production cycle analysis is carried out to the data after division, product production process information and production limit information are generated, and further initial parameter is obtained, initial parameter is imported into production planning model, and production planning data of multiple production factories is generated by the plan analysis of ant colony algorithm in cloud platform.In the application, data mining is carried out in the big data of production factory, and an ant colony algorithm model is constructed based on the production process, the ant colony algorithm model is limited in path selection based on energy consumption, equipment running time and resource allocation dimension, and an optimal production factory plan can be finally obtained when simulating the ant path.

Description

Technical Field

[0001] This invention relates to the field of cloud computing, and more specifically, to a cloud-based method and system for analyzing factory operation data. Background Technology

[0002] Cloud computing, big data, and other technologies hold a crucial position in the computer field, not only being widely applied across various sectors such as industry, healthcare, and artificial intelligence, but also representing a revolutionary technological advancement. Furthermore, with the widespread adoption of Industry 4.0 and smart manufacturing technologies, large-scale, group-based production models are gradually becoming the mainstream in manufacturing. Most manufacturing enterprises have established multi-regional, multi-plant collaborative production systems. During factory production, massive amounts of multi-dimensional production and operational big data are continuously generated. This data covers core information such as production processing, equipment operation, personnel scheduling, material supply, and energy consumption, serving as the core basis for enterprises to optimize production processes, reduce production costs, improve production efficiency, and achieve refined operations. Efficient and accurate data analysis and production planning directly determine the overall benefits and market competitiveness of multi-plant collaborative production.

[0003] However, in current factory production and large-scale manufacturing enterprises, the lack of cloud computing and matching model technology for real-time production data analysis makes it difficult to unlock data value and improve factory manufacturing efficiency. Furthermore, in current enterprise processing and production, the lack of accurate and efficient analysis of factory operational data results in a lack of comprehensive analysis and planning matching processes for factory processing workflows and the allocation of processing plans between various factories. This easily leads to problems such as chaotic data classification, data redundancy, and inaccurate selection of effective data, significantly reducing the accuracy of subsequent data analysis and production planning, further resulting in low production efficiency across factories and hindering innovative development. Summary of the Invention

[0004] This invention overcomes the shortcomings of existing technologies and proposes a cloud computing-based method and system for analyzing factory operation data.

[0005] The first aspect of this invention provides a cloud computing-based method for analyzing factory operation data, comprising: Acquire big data from multiple production plants; Based on the big data from the multiple production plants, a decision tree-based classification model is used to divide the data, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The product production cycle is analyzed by collecting product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and product production process information and production restriction information are generated based on multiple production plants. The product production process information is sent to the cloud platform, and a production planning model based on the ant colony algorithm is built on the cloud platform. Initial parameters are generated based on product production process information and production constraint information. These initial parameters are then imported into the production planning model for planning analysis based on the ant colony algorithm, and production plan data for multiple production plants are generated.

[0006] In this solution, obtaining big data from multiple production plants specifically involves: Acquire product production data, equipment operation data, and resource allocation data from multiple production plants; The product production data, equipment operation data, and resource allocation data are fused together to obtain initial big data from multiple production plants; The initial big data is cleaned and redundancy is removed to obtain production plant big data from multiple production plants.

[0007] In this solution, the step of dividing the data based on a decision tree-based classification model using the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data includes: Construct a classification model based on decision trees; Obtain standardized data based on product production, factory operations, equipment management, and resource allocation from the production plant database; Based on the overall production process of the production plant, hierarchical analysis is performed on the data of each production link to obtain hierarchical data information. The standardized data is subjected to feature extraction based on multiple dimensions such as production process, factory operation, equipment management, and resource allocation to obtain multiple types of production feature data; Each type of production characteristic data is transformed into decision conditions to obtain multiple corresponding decision nodes; Based on the hierarchical information of the data, the positions of multiple decision nodes are arranged, with the node corresponding to the main production feature data as the root node or parent node, and the decision node corresponding to the secondary production feature data as the parent node or leaf node, thereby forming a classification model with decision branching function. Randomly acquire a preset amount of historical production big data from multiple production plants; The historical production big data is imported into a classification model for data classification, and the classification result data is obtained. Calculate the data difference between the classification result data and the standardized data according to the standard Euclidean distance calculation method; Determine whether the data gap value is greater than a preset gap value. If it is, adjust the position distribution of decision nodes in the classification model and cyclically import the historical production big data into the classification model for data classification training until the final data gap value is less than or equal to the preset gap value.

[0008] In this solution, the data is divided using a decision tree-based classification model based on the big data from the multiple production plants, resulting in product production data, production operation data, equipment maintenance data, and human and material resource allocation data. Specifically: Import big data from multiple production plants into a classification model; Select a production plant's big data, standardize the data, and classify the data based on the dimensions of production, operation, equipment, and resources to obtain corresponding product production data, production operation data, equipment maintenance data, and human and material resource allocation data; The big data of the remaining production plants is classified to obtain the product production data, production operation data, equipment operation and maintenance data, and human and material resource allocation data corresponding to each production plant.

[0009] In this solution, the product production cycle is analyzed using product production data, production operation data, equipment maintenance data, and human and material resource allocation data. Based on multiple production plants, product production process information and production constraint information are generated, specifically as follows: Based on a selected production plant, the corresponding product production data and production operation data are analyzed cyclically to obtain product production process information. Obtain standard data on energy consumption and equipment operation in production plants; Based on the energy consumption standard data and equipment operation standard data of the production plant, the equipment operation and maintenance data are subjected to a limitation analysis based on energy consumption and operating time to obtain energy consumption limitation information and equipment time limitation information. Based on the aforementioned human and material resource allocation data, resource constraint analysis is performed to obtain human and material resource constraint information. The energy consumption restriction information, equipment time restriction information, and human and material resource restriction information are fused together to obtain production restriction information.

[0010] In this solution, the step of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically involves: Obtain overall task requirements information; Extract the overall task requirements information to identify the production task requirements of the selected production plant. Based on the product production process information of the selected production plant, an ant colony simulation structure is built from the cloud platform to form an initial simulation model based on a graph structure. Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges based on the corresponding graph structure in the initial simulation model. Based on the production task requirements, the starting point and ending point are set for the initial simulation model, and a production planning model for the selected production plant is formed.

[0011] In this solution, the process of generating initial parameters based on product production process information and production constraint information, importing these initial parameters into a production planning model, performing planning analysis based on the ant colony algorithm, and generating production plan data for multiple production plants specifically involves: Based on the product production process information and production task requirements of the selected production plant, analyze the scale of the production task, and calculate the number of ant colony test units m based on the scale of the task. Based on the production constraints of the selected production plant, and taking into account energy consumption, equipment uptime, and resource allocation, an initial pheromone value and a pheromone weight are generated. The ant colony test quantity m, the initial pheromone value, and the pheromone weight are integrated into the initial parameters and imported into the production planning model. Initialize the model parameters of the production planning model based on the initial parameters; Based on the number of ants m and the starting position, simulate the movement of the ant colony in the production planning model until m ants reach the destination or exceed the preset distance during the simulation. Repeatedly perform ant colony simulation movement tests, updating the pheromone based on the pheromone weight and ant colony path during each test, until the preset number of iterations is reached; In each ant colony simulated movement test, the shortest legal path for each test is recorded to obtain the optimal path information; Based on the optimal path information and product production process information, path tracing is performed in the production planning model to generate multiple production plan data. Analyze the remaining non-selected production plants and obtain multiple sets of corresponding production plan data.

[0012] A second aspect of the present invention also provides a cloud computing-based factory operation data analysis system, the system comprising: a memory and a processor, wherein the memory includes a cloud computing-based factory operation data analysis program, and the cloud computing-based factory operation data analysis program, when executed by the processor, performs the following steps: Acquire big data from multiple production plants; Based on the big data from the multiple production plants, a decision tree-based classification model is used to divide the data, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The product production cycle is analyzed by collecting product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and product production process information and production restriction information are generated based on multiple production plants. The product production process information is sent to the cloud platform, and a production planning model based on the ant colony algorithm is built on the cloud platform. Initial parameters are generated based on product production process information and production constraint information. These initial parameters are then imported into the production planning model for planning analysis based on the ant colony algorithm, and production plan data for multiple production plants are generated.

[0013] In this solution, obtaining big data from multiple production plants specifically involves: Acquire product production data, equipment operation data, and resource allocation data from multiple production plants; The product production data, equipment operation data, and resource allocation data are fused together to obtain initial big data from multiple production plants; The initial big data is cleaned and redundancy is removed to obtain production plant big data from multiple production plants.

[0014] In this solution, the step of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically involves: Obtain overall task requirements information; Extract the overall task requirements information to identify the production task requirements of the selected production plant. Based on the product production process information of the selected production plant, an ant colony simulation structure is built from the cloud platform to form an initial simulation model based on a graph structure. Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges based on the corresponding graph structure in the initial simulation model. Based on the production task requirements, the starting point and ending point are set for the initial simulation model, and a production planning model for the selected production plant is formed.

[0015] This invention discloses a cloud-based factory operation data analysis method and system. It acquires big data from multiple production plants and divides the data using a decision tree classification model, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The divided data is then analyzed for product production cycles to generate product production process information and production constraint information, and initial parameters are obtained. These initial parameters are imported into a production planning model, and ant colony optimization (ACO) is used in a cloud platform to generate production plan data for multiple production plants. In this invention, data mining is performed on the big data from the production plants, and an ACO model is constructed based on the production process. By imposing path selection constraints on the ACO model based on energy consumption, equipment uptime, and resource allocation dimensions, a superior production plant plan can be obtained during ant path simulation. Attached Figure Description

[0016] Figure 1 A flowchart of a cloud computing-based factory operation data analysis method according to the present invention is shown; Figure 2 This invention illustrates a flowchart of the process for acquiring big data from a production plant. Figure 3 The flowchart illustrating the production planning model construction of this invention is shown. Figure 4 A block diagram of a cloud-based factory operation data analysis system according to the present invention is shown. Detailed Implementation

[0017] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0018] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0019] Figure 1 A flowchart of a cloud computing-based factory operation data analysis method according to the present invention is shown.

[0020] like Figure 1 As shown, the first aspect of the present invention provides a cloud computing-based factory operation data analysis method, comprising: S102, acquires big data from multiple production plants; S104, Based on the big data of the multiple production plants, a decision tree-based classification model is used to divide the data and form product production data, production operation data, equipment operation and maintenance data, and human and material resource allocation data. S106 analyzes the product production cycle by combining product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and generates product production process information and production restriction information based on multiple production plants. S108 sends product production process information to the cloud platform and builds a production planning model based on the ant colony algorithm based on the cloud platform. S110 generates initial parameters based on product production process information and production constraint information, imports the initial parameters into the production planning model, performs planning analysis based on ant colony algorithm, and generates production plan data for multiple production plants.

[0021] It should be noted that in the method of this invention, the terminal platform includes multiple production plant terminal platforms and a cloud platform, and the platforms can achieve data connection based on the Internet. The production planning model is specifically built and analyzed on the cloud platform, further reducing the data analysis pressure on the factory terminal platforms. The cloud platform realizes data cloud computing and data interaction of planning results to each terminal by deploying the model.

[0022] Figure 2 The flowchart illustrating the process of acquiring big data from a production plant according to the present invention is shown.

[0023] According to an embodiment of the present invention, the acquisition of big data from multiple production plants specifically includes: S202, obtain product production data, equipment operation data, and resource allocation data from multiple production plants; S204, The product production data, equipment operation data, and resource allocation data are fused to obtain initial big data from multiple production plants; S206, perform data cleaning and redundancy removal operations on the initial big data to obtain production plant big data from multiple production plants.

[0024] It should be noted that the product production data, equipment operation data, and resource allocation data include both historical data and currently collected real-time data.

[0025] According to an embodiment of the present invention, the step of dividing the data based on a decision tree-based classification model using the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data includes: Construct a classification model based on decision trees; Obtain standardized data based on product production, factory operations, equipment management, and resource allocation from the production plant database; Based on the overall production process of the production plant, hierarchical analysis is performed on the data of each production link to obtain hierarchical data information. The standardized data is subjected to feature extraction based on multiple dimensions such as production process, factory operation, equipment management, and resource allocation to obtain multiple types of production feature data; Each type of production characteristic data is transformed into decision conditions to obtain multiple corresponding decision nodes; Based on the hierarchical information of the data, the positions of multiple decision nodes are arranged, with the node corresponding to the main production feature data as the root node or parent node, and the decision node corresponding to the secondary production feature data as the parent node or leaf node, thereby forming a classification model with decision branching function. Randomly acquire a preset amount of historical production big data from multiple production plants; The historical production big data is imported into a classification model for data classification, and the classification result data is obtained. Calculate the data difference between the classification result data and the standardized data according to the standard Euclidean distance calculation method; Determine whether the data gap value is greater than a preset gap value. If it is, adjust the position distribution of decision nodes in the classification model and cyclically import the historical production big data into the classification model for data classification training until the final data gap value is less than or equal to the preset gap value.

[0026] It should be noted that the standardized data specifically refers to production, operation, and resource allocation data with standard data formats and representativeness, selected by the user, and possessing high feature density. The data hierarchy information specifically refers to the hierarchical relationships of data in product production, factory operation, equipment management, and resource allocation within the enterprise's production process, used to describe which data is primary, secondary, or dependent. Through data hierarchy information, a basic decision tree can be effectively constructed, improving the efficiency of initial decision-making. The historical production big data includes production data based on product production, factory operation, equipment management, and resource allocation. Adjusting the distribution of decision nodes in the classification model requires ensuring that nodes corresponding to primary production feature data are used as root nodes or parent nodes, and decision nodes corresponding to secondary production feature data are used as parent nodes or leaf nodes.

[0027] According to an embodiment of the present invention, the step of dividing the data based on a decision tree-based classification model using the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data specifically includes: Import big data from multiple production plants into a classification model; Select a production plant's big data, standardize the data, and classify the data based on the dimensions of production, operation, equipment, and resources to obtain corresponding product production data, production operation data, equipment maintenance data, and human and material resource allocation data; The big data of the remaining production plants is classified to obtain the product production data, production operation data, equipment operation and maintenance data, and human and material resource allocation data corresponding to each production plant.

[0028] It's worth noting that the categorized data obtained through the classification model can be used for accurate ant colony modeling in subsequent analysis, thereby yielding efficient planning data. Furthermore, the classification model enables the rapid classification of large datasets, improving system operational efficiency.

[0029] According to an embodiment of the present invention, the step of performing product production cycle analysis on product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and generating product production process information and production restriction information based on multiple production plants, specifically includes: Based on a selected production plant, the corresponding product production data and production operation data are analyzed cyclically to obtain product production process information. Obtain standard data on energy consumption and equipment operation in production plants; Based on the energy consumption standard data and equipment operation standard data of the production plant, the equipment operation and maintenance data are subjected to a limitation analysis based on energy consumption and operating time to obtain energy consumption limitation information and equipment time limitation information. Based on the aforementioned human and material resource allocation data, resource constraint analysis is performed to obtain human and material resource constraint information. The energy consumption restriction information, equipment time restriction information, and human and material resource restriction information are fused together to obtain production restriction information.

[0030] It should be noted that, through the analysis process of this invention, information on the product production process and production restrictions corresponding to each production plant can be obtained.

[0031] Figure 3 The flowchart illustrating the production planning model construction of this invention is shown.

[0032] According to an embodiment of the present invention, the step of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically includes: S302, Obtain overall task requirements information; S304, extract the overall task requirement information from the production task requirements of the selected production plant; S306: Based on the product production process information of the selected production plant, build the ant colony simulation structure from the cloud platform to form an initial simulation model based on the graph structure. S308: Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges in the graph structure in the initial simulation model. S310, Based on the production task requirements, the starting point and ending point of the initial simulation model are set, and a production planning model for the selected production plant is formed.

[0033] It should be noted that the product production process structure diagram includes the structural flow diagram of the production processes, equipment, and personnel allocation in the production process. This diagram data can be converted into relevant data based on computer graph structures. When extracting the production task requirements of a selected production plant from the overall task requirement information, since the overall production plan of a product often involves the processing of multiple production plants to obtain the final product, extracting and analyzing the production task requirements of a specific production plant from the overall task requirement information allows for a refined analysis of the entire product production cycle. In the production planning model, different production plants correspond to different production planning models. In the initial simulation model, the vertices and edges generated based on the graph structure correspond to the process nodes of the factory production, and the edges correspond to the subdivided processes of the factory production.

[0034] According to an embodiment of the present invention, the step of generating initial parameters based on product production process information and production constraint information, importing the initial parameters into a production planning model, performing planning analysis based on ant colony algorithm, and generating production plan data for multiple production plants specifically includes: Based on the product production process information and production task requirements of the selected production plant, analyze the scale of the production task, and calculate the number of ant colony test units m based on the scale of the task. Based on the production constraints of the selected production plant, and taking into account energy consumption, equipment uptime, and resource allocation, an initial pheromone value and a pheromone weight are generated. The ant colony test quantity m, the initial pheromone value, and the pheromone weight are integrated into the initial parameters and imported into the production planning model. Initialize the model parameters of the production planning model based on the initial parameters; Based on the number of ants m and the starting position, simulate the movement of the ant colony in the production planning model until m ants reach the destination or exceed the preset distance during the simulation. Repeatedly perform ant colony simulation movement tests, updating the pheromone based on the pheromone weight and ant colony path during each test, until the preset number of iterations is reached; In each ant colony simulated movement test, the shortest legal path for each test is recorded to obtain the optimal path information; Based on the optimal path information and product production process information, path tracing is performed in the production planning model to generate multiple production plan data. Analyze the remaining non-selected production plants and obtain multiple sets of corresponding production plan data.

[0035] It should be noted that in the calculation of the number of ant colony tests *m* based on the task size, the task size is directly proportional to *m*; the larger the task size, the larger *m*. The number of ant colony tests *m* refers to the number of ants in each test. The valid path is the effectively recorded path of the ants from the starting point to the ending point. The production planning model includes all key information of the product production process. Through effective tracing of valid paths, the corresponding production process information can be linked together in the production planning model to form production plan data. The larger the pheromone weight, the smaller the pheromone reduction on the corresponding path during the ant colony model testing iteration process.

[0036] In this invention, by mining data from big data in the production plant and constructing an ant colony algorithm model based on the production process, and by imposing path selection restrictions on the ant colony algorithm model based on energy consumption, equipment uptime, and resource allocation dimensions, a better production plant plan can be obtained when ant path simulation is performed.

[0037] According to an embodiment of the present invention, it further includes: Obtain multiple production plan data sets from a selected production plant; Each production plan data is comprehensively calculated and analyzed based on energy consumption, equipment, and human and material resources to obtain the corresponding resource consumption index; Based on the overall task requirements information, the production task proportion in the selected production plant is analyzed. The initial pheromone is calculated based on the production task proportion. The pheromone weights are calculated based on the resource consumption index. Each piece of production plan data is treated as a path point, and the initial pheromone and pheromone weight corresponding to the production plan data are used as the pheromone parameters of that path point. The path points of the remaining non-selected production plants were analyzed, and a simulation model was built for all path points based on the ant colony algorithm to obtain the ant colony model. Set preset movement rules for the ant colony model; The number of ant colony tests, m, is set based on the total number of path points. Based on the number of ant colony tests m and the ant colony model, simulate the ant colony path movement and record the second legal path. The ant colony path movement simulation is repeated until the preset number of iterations is reached, resulting in multiple second legal paths; Based on the multiple second legal paths, path point tracing is performed, and data integration is performed based on the production plan data corresponding to the path points to obtain multiple overall production plans.

[0038] It should be noted that within a selected production plant, due to the fixed proportion of production tasks, the initial pheromone for all path points corresponding to the production plan data of that production plant is the same, but the pheromone weights differ depending on the resource consumption index. One set of production plan data corresponds to one path point. The second valid path must include path points from all production plants. The initial pheromone is obtained by multiplying a preset pheromone correction value by the production task proportion, and the pheromone weight is obtained by multiplying the preset pheromone correction value by the resource consumption index.

[0039] The preset movement rule stipulates that each movement of an ant in the ant colony requires moving from a waypoint in one production factory to a waypoint in another production factory. An overall production plan includes N sets of production plan data, where each set of production plan data corresponds to one production factory, and N is the number of production factories. One second valid path corresponds to one overall production plan.

[0040] The present invention enables ant colony algorithm modeling based on the planning data of all production plants. It uses the production task ratio and resource consumption index as reference values ​​for relevant information pheromone parameters to perform simulation and combination analysis of the planning data, and finally obtains the preferred second legal path. The second legal path is used to trace the data source and form an overall production plan, thereby analyzing the optimal overall production allocation plan among multiple production plants, and achieving cost reduction and efficiency improvement for enterprise production.

[0041] Figure 4 A block diagram of a cloud-based factory operation data analysis system according to the present invention is shown.

[0042] A second aspect of the present invention also provides a cloud computing-based factory operation data analysis system 4, which includes: a memory 41 and a processor 42. The memory includes a cloud computing-based factory operation data analysis program, which, when executed by the processor, performs the following steps: Acquire big data from multiple production plants; Based on the big data from the multiple production plants, a decision tree-based classification model is used to divide the data, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The product production cycle is analyzed by collecting product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and product production process information and production restriction information are generated based on multiple production plants. The product production process information is sent to the cloud platform, and a production planning model based on the ant colony algorithm is built on the cloud platform. Initial parameters are generated based on product production process information and production constraint information. These initial parameters are then imported into the production planning model for planning analysis based on the ant colony algorithm, and production plan data for multiple production plants are generated.

[0043] It should be noted that in the method of this invention, the terminal platform includes multiple production plant terminal platforms and a cloud platform, and the platforms can achieve data connection based on the Internet. Specifically, the production planning model is built and data analyzed on the cloud platform, further reducing the data analysis burden on the factory terminal platforms.

[0044] According to an embodiment of the present invention, the acquisition of big data from multiple production plants specifically includes: Acquire product production data, equipment operation data, and resource allocation data from multiple production plants; The product production data, equipment operation data, and resource allocation data are fused together to obtain initial big data from multiple production plants; The initial big data is cleaned and redundancy is removed to obtain production plant big data from multiple production plants.

[0045] It should be noted that the product production data, equipment operation data, and resource allocation data include both historical data and currently collected real-time data.

[0046] According to an embodiment of the present invention, the step of dividing the data based on a decision tree-based classification model using the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data includes: Construct a classification model based on decision trees; Obtain standardized data based on product production, factory operations, equipment management, and resource allocation from the production plant database; Based on the overall production process of the production plant, hierarchical analysis is performed on the data of each production link to obtain hierarchical data information. The standardized data is subjected to feature extraction based on multiple dimensions such as production process, factory operation, equipment management, and resource allocation to obtain multiple types of production feature data; Each type of production characteristic data is transformed into decision conditions to obtain multiple corresponding decision nodes; Based on the hierarchical information of the data, the positions of multiple decision nodes are arranged, with the node corresponding to the main production feature data as the root node or parent node, and the decision node corresponding to the secondary production feature data as the parent node or leaf node, thereby forming a classification model with decision branching function. Randomly acquire a preset amount of historical production big data from multiple production plants; The historical production big data is imported into a classification model for data classification, and the classification result data is obtained. Calculate the data difference between the classification result data and the standardized data according to the standard Euclidean distance calculation method; Determine whether the data gap value is greater than a preset gap value. If it is, adjust the position distribution of decision nodes in the classification model and cyclically import the historical production big data into the classification model for data classification training until the final data gap value is less than or equal to the preset gap value.

[0047] It should be noted that the standardized data specifically refers to production, operation, and resource allocation data with standard data formats and representativeness, selected by the user, and possessing high feature density. The data hierarchy information specifically refers to the hierarchical relationships of data in product production, factory operation, equipment management, and resource allocation within the enterprise's production process, used to describe which data is primary, secondary, or dependent. Through data hierarchy information, a basic decision tree can be effectively constructed, improving the efficiency of initial decision-making. The historical production big data includes production data based on product production, factory operation, equipment management, and resource allocation. Adjusting the distribution of decision nodes in the classification model requires ensuring that nodes corresponding to primary production feature data are used as root nodes or parent nodes, and decision nodes corresponding to secondary production feature data are used as parent nodes or leaf nodes.

[0048] According to an embodiment of the present invention, the step of dividing the data based on a decision tree-based classification model using the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data specifically includes: Import big data from multiple production plants into a classification model; Select a production plant's big data, standardize the data, and classify the data based on the dimensions of production, operation, equipment, and resources to obtain corresponding product production data, production operation data, equipment maintenance data, and human and material resource allocation data; The big data of the remaining production plants is classified to obtain the product production data, production operation data, equipment operation and maintenance data, and human and material resource allocation data corresponding to each production plant.

[0049] It's worth noting that the categorized data obtained through the classification model can be used for accurate ant colony modeling in subsequent analysis, thereby yielding efficient planning data. Furthermore, the classification model enables the rapid classification of large datasets, improving system operational efficiency.

[0050] According to an embodiment of the present invention, the step of performing product production cycle analysis on product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and generating product production process information and production restriction information based on multiple production plants, specifically includes: Based on a selected production plant, the corresponding product production data and production operation data are analyzed cyclically to obtain product production process information. Obtain standard data on energy consumption and equipment operation in production plants; Based on the energy consumption standard data and equipment operation standard data of the production plant, the equipment operation and maintenance data are subjected to a limitation analysis based on energy consumption and operating time to obtain energy consumption limitation information and equipment time limitation information. Based on the aforementioned human and material resource allocation data, resource constraint analysis is performed to obtain human and material resource constraint information. The energy consumption restriction information, equipment time restriction information, and human and material resource restriction information are fused together to obtain production restriction information.

[0051] It should be noted that, through the analysis process of this invention, information on the product production process and production restrictions corresponding to each production plant can be obtained.

[0052] According to an embodiment of the present invention, the step of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically includes: Obtain overall task requirements information; Extract the overall task requirements information to identify the production task requirements of the selected production plant. Based on the product production process information of the selected production plant, an ant colony simulation structure is built from the cloud platform to form an initial simulation model based on a graph structure. Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges based on the corresponding graph structure in the initial simulation model. Based on the production task requirements, the starting point and ending point are set for the initial simulation model, and a production planning model for the selected production plant is formed.

[0053] It should be noted that the product production process structure diagram includes the structural flow diagram of the production processes, equipment, and personnel allocation in the production process. This diagram data can be converted into relevant data based on computer graph structures. When extracting the production task requirements of a selected production plant from the overall task requirement information, since the overall production plan of a product often involves the processing of multiple production plants to obtain the final product, extracting and analyzing the production task requirements of a specific production plant from the overall task requirement information allows for a refined analysis of the entire product production cycle. In the production planning model, different production plants correspond to different production planning models. In the initial simulation model, the vertices and edges generated based on the graph structure correspond to the process nodes of the factory production, and the edges correspond to the subdivided processes of the factory production.

[0054] According to an embodiment of the present invention, the step of generating initial parameters based on product production process information and production constraint information, importing the initial parameters into a production planning model, performing planning analysis based on ant colony algorithm, and generating production plan data for multiple production plants specifically includes: Based on the product production process information and production task requirements of the selected production plant, analyze the scale of the production task, and calculate the number of ant colony test units m based on the scale of the task. Based on the production constraints of the selected production plant, and taking into account energy consumption, equipment uptime, and resource allocation, an initial pheromone value and a pheromone weight are generated. The ant colony test quantity m, the initial pheromone value, and the pheromone weight are integrated into the initial parameters and imported into the production planning model. Initialize the model parameters of the production planning model based on the initial parameters; Based on the number of ants m and the starting position, simulate the movement of the ant colony in the production planning model until m ants reach the destination or exceed the preset distance during the simulation. Repeatedly perform ant colony simulation movement tests, updating the pheromone based on the pheromone weight and ant colony path during each test, until the preset number of iterations is reached; In each ant colony simulated movement test, the shortest legal path for each test is recorded to obtain the optimal path information; Based on the optimal path information and product production process information, path tracing is performed in the production planning model to generate multiple production plan data. Analyze the remaining non-selected production plants and obtain multiple sets of corresponding production plan data.

[0055] It should be noted that in the calculation of the number of ant colony tests *m* based on the task size, the task size is directly proportional to *m*; the larger the task size, the larger *m*. The number of ant colony tests *m* refers to the number of ants in each test. The valid path is the effectively recorded path of the ants from the starting point to the ending point. The production planning model includes all key information of the product production process. Through effective tracing of valid paths, the corresponding production process information can be linked together in the production planning model to form production plan data. The larger the pheromone weight, the smaller the pheromone reduction on the corresponding path during the ant colony model testing iteration process.

[0056] In this invention, by mining data from big data in the production plant and constructing an ant colony algorithm model based on the production process, and by imposing path selection restrictions on the ant colony algorithm model based on energy consumption, equipment uptime, and resource allocation dimensions, a better production plant plan can be obtained when ant path simulation is performed.

[0057] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0058] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0059] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0060] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0061] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0062] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A cloud computing-based factory operation data analysis method, characterized in that, include: Acquire big data from multiple production plants; Based on the big data from the multiple production plants, a decision tree-based classification model is used to divide the data, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The product production cycle is analyzed by collecting product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and product production process information and production restriction information are generated based on multiple production plants. The product production process information is sent to the cloud platform, and a production planning model based on the ant colony algorithm is built on the cloud platform. Initial parameters are generated based on product production process information and production constraint information. These initial parameters are then imported into the production planning model for planning analysis based on the ant colony algorithm, and production plan data for multiple production plants are generated.

2. The factory operation data analysis method based on cloud computing according to claim 1, characterized in that, The acquisition of big data from multiple production plants specifically includes: Acquire product production data, equipment operation data, and resource allocation data from multiple production plants; The product production data, equipment operation data, and resource allocation data are fused together to obtain initial big data from multiple production plants; The initial big data is cleaned and redundancy is removed to obtain production plant big data from multiple production plants.

3. The factory operation data analysis method based on cloud computing according to claim 1, characterized in that, The process of dividing the data using a decision tree-based classification model based on the big data from the multiple production plants to form product production data, production operation data, equipment maintenance data, and human and material resource allocation data includes: Construct a classification model based on decision trees; Obtain standardized data based on product production, factory operations, equipment management, and resource allocation from the production plant database; Based on the overall production process of the production plant, hierarchical analysis is performed on the data of each production link to obtain hierarchical data information. The standardized data is subjected to feature extraction based on multiple dimensions such as production process, factory operation, equipment management, and resource allocation to obtain multiple types of production feature data; Each type of production characteristic data is transformed into decision conditions to obtain multiple corresponding decision nodes; Based on the hierarchical information of the data, the positions of multiple decision nodes are arranged, with the node corresponding to the main production feature data as the root node or parent node, and the decision node corresponding to the secondary production feature data as the parent node or leaf node, thereby forming a classification model with decision branching function. Randomly acquire a preset amount of historical production big data from multiple production plants; The historical production big data is imported into a classification model for data classification, and the classification result data is obtained. Calculate the data difference between the classification result data and the standardized data according to the standard Euclidean distance calculation method; Determine whether the data gap value is greater than a preset gap value. If it is, adjust the position distribution of decision nodes in the classification model and cyclically import the historical production big data into the classification model for data classification training until the final data gap value is less than or equal to the preset gap value.

4. The factory operation data analysis method based on cloud computing according to claim 3, characterized in that, The process involves using a decision tree-based classification model to segment the data based on the big data from the multiple production plants, resulting in product production data, production operation data, equipment maintenance data, and human and material resource allocation data. Specifically: Import big data from multiple production plants into a classification model; Select a production plant's big data, standardize the data, and classify the data based on the dimensions of production, operation, equipment, and resources to obtain corresponding product production data, production operation data, equipment maintenance data, and human and material resource allocation data; The big data of the remaining production plants is classified to obtain the product production data, production operation data, equipment operation and maintenance data, and human and material resource allocation data corresponding to each production plant.

5. The factory operation data analysis method based on cloud computing according to claim 1, characterized in that, The process involves analyzing product production data, production operation data, equipment maintenance data, and human and material resource allocation data to determine the product production cycle. Based on multiple production plants, it generates product production process information and production constraint information. Specifically: Based on a selected production plant, the corresponding product production data and production operation data are analyzed cyclically to obtain product production process information. Obtain standard data on energy consumption and equipment operation in production plants; Based on the energy consumption standard data and equipment operation standard data of the production plant, the equipment operation and maintenance data are subjected to a limitation analysis based on energy consumption and operating time to obtain energy consumption limitation information and equipment time limitation information. Based on the aforementioned human and material resource allocation data, resource constraint analysis is performed to obtain human and material resource constraint information. The energy consumption restriction information, equipment time restriction information, and human and material resource restriction information are fused together to obtain production restriction information.

6. The factory operation data analysis method based on cloud computing according to claim 5, characterized in that, The process of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically involves: Obtain overall task requirements information; Extract the overall task requirements information to identify the production task requirements of the selected production plant. Based on the product production process information of the selected production plant, an ant colony simulation structure is built from the cloud platform to form an initial simulation model based on a graph structure. Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges based on the corresponding graph structure in the initial simulation model. Based on the production task requirements, the starting point and ending point are set for the initial simulation model, and a production planning model for the selected production plant is formed.

7. The factory operation data analysis method based on cloud computing according to claim 6, characterized in that, The process involves generating initial parameters based on product production process information and production constraint information, importing these initial parameters into a production planning model, performing planning analysis based on the ant colony algorithm, and generating production plan data for multiple production plants. Specifically: Based on the product production process information and production task requirements of the selected production plant, analyze the scale of the production task, and calculate the number of ant colony test units m based on the scale of the task. Based on the production constraints of the selected production plant, and taking into account energy consumption, equipment uptime, and resource allocation, an initial pheromone value and a pheromone weight are generated. The ant colony test quantity m, the initial pheromone value, and the pheromone weight are integrated into the initial parameters and imported into the production planning model. Initialize the model parameters of the production planning model based on the initial parameters; Based on the number of ants m and the starting position, simulate the movement of the ant colony in the production planning model until m ants reach the destination or exceed the preset distance during the simulation. Repeatedly perform ant colony simulation movement tests, updating the pheromone based on the pheromone weight and ant colony path during each test, until the preset number of iterations is reached; In each ant colony simulated movement test, the shortest legal path for each test is recorded to obtain the optimal path information; Based on the optimal path information and product production process information, path tracing is performed in the production planning model to generate multiple production plan data. Analyze the remaining non-selected production plants and obtain multiple sets of corresponding production plan data.

8. A cloud computing-based factory operation data analysis system, characterized in that, The system includes: a memory and a processor. The memory includes a cloud-based factory operation data analysis program. When the cloud-based factory operation data analysis program is executed by the processor, it performs the following steps: Acquire big data from multiple production plants; Based on the big data from the multiple production plants, a decision tree-based classification model is used to divide the data, forming product production data, production operation data, equipment maintenance data, and human and material resource allocation data. The product production cycle is analyzed by collecting product production data, production operation data, equipment maintenance data, and human and material resource allocation data, and product production process information and production restriction information are generated based on multiple production plants. The product production process information is sent to the cloud platform, and a production planning model based on the ant colony algorithm is built on the cloud platform. Initial parameters are generated based on product production process information and production constraint information. These initial parameters are then imported into the production planning model for planning analysis based on the ant colony algorithm, and production plan data for multiple production plants are generated.

9. A cloud-based factory operation data analysis system according to claim 8, characterized in that, The acquisition of big data from multiple production plants specifically includes: Acquire product production data, equipment operation data, and resource allocation data from multiple production plants; The product production data, equipment operation data, and resource allocation data are fused together to obtain initial big data from multiple production plants; The initial big data is cleaned and redundancy is removed to obtain production plant big data from multiple production plants.

10. A cloud-based factory operation data analysis system according to claim 8, characterized in that, The process of sending product production process information to a cloud platform and constructing a production planning model based on the ant colony algorithm on the cloud platform specifically involves: Obtain overall task requirements information; Extract the overall task requirements information to identify the production task requirements of the selected production plant. Based on the product production process information of the selected production plant, an ant colony simulation structure is built from the cloud platform to form an initial simulation model based on a graph structure. Generate a product production process structure diagram based on the product production process information, perform structural mapping on the initial simulation model based on the product production process structure diagram, and generate vertices and edges based on the corresponding graph structure in the initial simulation model. Based on the production task requirements, the starting point and ending point are set for the initial simulation model, and a production planning model for the selected production plant is formed.

Images