A production plan evaluation method and system for aviation manufacturing industry
By adopting production planning evaluation methods in the aviation manufacturing industry, using technologies such as reinforcement learning and optimized data grouping algorithms, the problem of lack of real-time data support in the existing technology has been solved, and the refinement and dynamic adjustment of production plans has been achieved, and production has been improved. Efficiency and product quality.
Patent Information
- Application Number
- CN202411270489.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-11
AI Technical Summary
The lack of decision-making support based on real-time data and advanced algorithms in the aviation manufacturing industry, resulting in difficulty in optimizing production plans and inability to adjust in time, affecting production efficiency and product quality, especially when facing complex and varied production requirements.
The production plan evaluation method is adopted, including formulating overall goals, designing specific implementation plans, simulation and data processing, establishing a dynamic monitoring mechanism, using reinforcement learning algorithms and optimized data grouping algorithms, adjusting production plans through clustering analysis and cuckoo search algorithms, ensuring that The production line performance is optimal.
The refinement and dynamic adjustment of production plans have been achieved, production efficiency and product quality have been improved, decision-making accuracy and effectiveness have been enhanced, problems in the production process can be identified and solved in a timely manner, and production line performance can be continuously optimized.
Smart Images

Figure CN118780648B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of aviation manufacturing industry, and in particular, to a production plan evaluation method and system for aviation manufacturing industry. Background Art
[0002] Aviation, as a broad field, focuses on the design, manufacture, operation and maintenance of aircraft. It includes not only commercial flights and military aircraft, but also drones, private aircraft and spacecraft for space exploration. The development of the aviation industry has promoted technological progress and social changes in many aspects of human beings, such as air travel, cargo transportation and space exploration.
[0003] The aviation manufacturing industry is a vital part of the aviation field, focusing on the design and manufacture of aircraft and their components. This industry requires extremely high technical standards and strict quality control, because any small mistake may lead to serious consequences. The aviation manufacturing industry covers the manufacture of various aircraft from small drones to large commercial aircraft and military aircraft. The manufacture of each aircraft requires precise engineering design, complex assembly process and detailed safety testing.
[0004] In the aviation manufacturing industry, the optimization and management of production plans are particularly important. It is not only related to the efficiency and cost control of the manufacturing process, but also directly affects the quality and delivery time of the product. A good production plan can ensure the effective use of resources and reduce the production cycle. At the same time, it can ensure product quality and safety, meet market demand and customer delivery deadlines. Therefore, aviation manufacturing companies usually adopt a highly systematic and scientific approach to formulate and implement production plans to achieve the above goals.
[0005] In the aviation manufacturing industry, the requirements for product quality are extremely strict. The production plan needs to ensure that each step meets the quality standards. From material procurement to final assembly, there must be no omissions in every link. Aviation manufacturing is a capital-intensive industry. By optimizing the production plan and reasonably arranging the production process and material use, costs can be effectively controlled and production efficiency can be improved. On-time delivery is the key to maintaining customer trust and satisfaction in the aviation manufacturing industry. Reasonable production plans can ensure a smooth production process, avoid delays, and ensure on-time delivery of products.
[0006] Existing technologies may fail to fully identify and adjust problems and inefficient links in the production process, which will affect overall production efficiency and cost-effectiveness. There is a lack of effective tools and methods to accurately evaluate and optimize the performance of the production line. In a rapidly changing market environment, timely adjustment of production plans to meet changes in market demand and raw material supply is a major challenge. Existing technologies may lack flexibility and responsiveness. In many production plans, there may be a lack of decision support based on large amounts of real-time data and advanced algorithms, which will affect the accuracy and effectiveness of decisions and limit the possibility of production optimization. There is a lack of effective mechanisms to monitor and adjust production plans in real time, which may result in the performance of the production line not being optimal and the inability to promptly resolve problems arising in the production process. Existing production methods may not be able to effectively improve production efficiency and ensure product quality, especially when faced with complex and changing production requirements.
[0007] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0008] In order to overcome the above problems, the present invention aims to propose a production plan evaluation method and system for the aviation manufacturing industry, with the aim of solving the problem that in many production plans, there may be a lack of decision support based on a large amount of real-time data and advanced algorithms, which affects the accuracy and effectiveness of decisions, limits the possibility of production optimization, and lacks an effective mechanism to monitor and adjust production plans in real time, which may lead to the performance of the production line not being optimal, and the inability to promptly resolve problems arising in the production process, and the existing production methods may not be able to effectively improve production efficiency and ensure product quality, especially when faced with complex and changing production requirements.
[0009] To this end, the specific technical solution adopted by the present invention is as follows:
[0010] According to one aspect of the present invention, a production plan evaluation method for aviation manufacturing is provided, the production plan evaluation method comprising the following steps:
[0011] S1. Formulate the overall goal of the aviation manufacturing industry production plan and construct evaluation indicators that reflect the effectiveness of the production plan execution;
[0012] S2. Design a specific implementation plan for each production line based on its characteristics and overall goals;
[0013] S3, simulating the implementation plan of each production line, and processing the data of each production line by optimizing the data grouping algorithm, and calculating the actual index value of the evaluation index of each production line;
[0014] S4. Establish a dynamic monitoring mechanism and use reinforcement learning algorithms to analyze the data of each production line;
[0015] S5. Compare the target value in the evaluation index with the actual index value to evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry based on the analysis results of the reinforcement learning algorithm.
[0016] Optionally, the overall objectives include quality objectives, cost objectives, and delivery objectives;
[0017] Evaluation indicators include production efficiency indicators, quality control indicators and cost control indicators.
[0018] Optionally, according to the characteristics and overall objectives of the production line, designing a specific implementation plan for each production line includes the following steps:
[0019] S21. Set specific quality goals, cost goals and delivery goals for each production line based on the quality goals, cost goals and delivery goals in the overall goals and the characteristics of each production line's current production capacity, technical capabilities, staffing and resource status;
[0020] S22. Develop quality control measures including quality inspection process, quality control points and quality improvement plan according to the quality objectives set for each production line;
[0021] S23. Analyze the cost structure of each production line, identify raw material costs and energy consumption, and optimize the supply chain according to the cost targets set for each production line;
[0022] S24. Develop a delivery schedule based on the delivery targets set for each production line, ensure that the production plan matches demand and market supply, implement supply chain collaboration, and optimize management processes.
[0023] Optionally, simulating each production line implementation plan, and processing the data of each production line by optimizing the data grouping algorithm, and calculating the actual index value of each production line evaluation index includes the following steps:
[0024] S31. For each production line, collect production data on production efficiency indicators, quality control indicators and cost control indicators;
[0025] S32. Based on the production data of each production line, simulation is performed using the implementation plan of the set quality target, cost target and delivery target, and simulated production line data is generated;
[0026] S33, using an optimized data grouping algorithm to perform cluster analysis on the simulated production line data;
[0027] S34, using the cuckoo search algorithm to iteratively optimize the production line data after cluster analysis, and finally select the clustering result with the highest fitness;
[0028] S35. Use the clustering results to identify abnormal patterns in the operation of the production line and analyze the strengths and weaknesses of each production line;
[0029] S36. Based on the results of cluster analysis, the actual index values of each production line in terms of production efficiency index, quality control index and cost control index are calculated.
[0030] Optionally, performing cluster analysis on the simulated production line data using an optimized data grouping algorithm comprises the following steps:
[0031] S331. Based on the collected production line data, select clustered features including production efficiency, quality control and cost control indicators;
[0032] S332, initializing the parameters of the optimized data grouping algorithm to generate an initial population, wherein each individual in the initial population represents a parameter set of a Gaussian mixture model;
[0033] S333, using each individual in the initial population to attempt to cluster the production line data;
[0034] S334, using the maximum likelihood function to calculate the fitness of each individual in the initial population;
[0035] Among them, the formula for calculating the fitness of each individual in the initial population using the maximum likelihood function is:
[0036] ;
[0037] In the formula, Represents the fitness of the parameter set in the Gaussian mixture model;
[0038] n represents the total number of data points in the parameter set;
[0039] Indicates the first i data points;
[0040] K represents the number of Gaussian distributions;
[0041] Indicates k A mixture of Gaussian distributions with weights;
[0042] Indicates k The mean vector of the Gaussian distribution;
[0043] Indicates k The covariance matrix of the Gaussian distribution;
[0044] represents the probability density function of the Gaussian distribution;
[0045] i means Data points in the parameter set;
[0046] k represents Gaussian distribution.
[0047] Optionally, using a cuckoo search algorithm to iteratively optimize the production line data after cluster analysis, and finally selecting the clustering result with the highest fitness includes the following steps:
[0048] S341, randomly select individuals in the population as cuckoos, and generate new solutions through Levy flight;
[0049] S342, using the maximum likelihood function to calculate the fitness of the newly generated solution;
[0050] S343. If the fitness of the newly generated solution is greater than the fitness of the cuckoo individual selected as the parent, or if it is randomly selected for replacement according to the discovery probability, the current solution is replaced by the new solution;
[0051] S344. In each iteration, the optimal solution encountered is recorded and updated, and the optimal solution recorded during the iteration process is used as the final clustering model parameter.
[0052] Optionally, based on the result of cluster analysis, calculating the actual index value of each production line in terms of production efficiency index, quality control index and cost control index comprises the following steps:
[0053] S361, obtaining the result of cluster analysis, and classifying each production line into a corresponding cluster category;
[0054] S362. For each cluster, calculate the average values of production efficiency index, quality control index and cost control index of all production lines in each cluster category;
[0055] S363, mapping the average value in the clustering result to each production line as the actual index value of each production line on the corresponding index;
[0056] The average calculation formula for production efficiency index, quality control index and cost control index is:
[0057] ;
[0058] ;
[0059] ;
[0060] In the formula, Indicates j The number of production lines in a cluster;
[0061] Indicates j Cluster categories;
[0062] Indicates j The average production efficiency index of each cluster category;
[0063] It represents the production efficiency index;
[0064] Indicates j The average quality control index of each cluster category;
[0065] Indicates quality control indicators;
[0066] Indicates j The average cost control index of each cluster category;
[0067] Represents cost control indicators.
[0068] Optionally, comparing the target value in the evaluation index with the actual index value, evaluating the overall effect of the production plan, and combining the analysis results of the reinforcement learning algorithm, adjusting the production plan of the aviation manufacturing industry includes the following steps:
[0069] S51, collecting actual operation data of each production line and actual index values of evaluation indexes;
[0070] S52, matching the actual index value with the target value set in the production plan and comparing them;
[0071] S53. Evaluate the overall execution effect of the production plan based on the comparison results;
[0072] S54, obtaining the result of reinforcement learning algorithm analysis;
[0073] S55. Based on the overall execution effect of the evaluated production plan and the results of the reinforcement learning algorithm analysis, identify the problems in the execution of the production plan and adjust the production plan of the aviation manufacturing industry.
[0074] Optionally, based on the comparison result, evaluating the overall execution effect of the production plan includes the following steps:
[0075] S531, analyzing the difference between the actual indicator value of each evaluation indicator and the preset target value;
[0076] S532. Based on the analysis results of the differences, identify the performance deficiencies in the production plan execution;
[0077] S533. For each aspect of insufficient performance, analyze the reasons why the actual indicators did not reach the target value;
[0078] S534. Integrate the actual performance of all evaluation indicators, evaluate the overall execution effect of the production plan, and summarize the overall performance of the production plan execution.
[0079] According to another aspect of the present invention, there is also provided a production plan evaluation system for the aviation manufacturing industry, the system comprising: a target setting and evaluation index construction module, a production line scheme design module, a simulation and data processing module, a dynamic monitoring module and an evaluation and plan adjustment module;
[0080] The target setting and evaluation index construction module is connected with the simulation and data processing module through the production line scheme design module, and the simulation and data processing module is connected with the evaluation and plan adjustment module through the dynamic monitoring module;
[0081] The module for goal setting and evaluation index construction is used to formulate the overall goal of the aviation manufacturing production plan and construct evaluation indicators that reflect the effect of production plan execution;
[0082] Production line solution design module, used to design specific implementation plans for each production line based on the characteristics and overall goals of the production line;
[0083] The simulation and data processing module is used to simulate the implementation plan of each production line, process the data of each production line by optimizing the data grouping algorithm, and calculate the actual index value of the evaluation index of each production line;
[0084] Dynamic monitoring module, which is used to establish a dynamic monitoring mechanism and analyze the data of each production line using reinforcement learning algorithms;
[0085] The evaluation and plan adjustment module is used to compare the target value in the evaluation index with the actual index value, evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry based on the analysis results of the reinforcement learning algorithm.
[0086] Compared with the prior art, this application has the following beneficial effects:
[0087] 1. The present invention sets quality, cost and delivery targets to provide clear guidance for production activities, help various departments focus on common goals, enhance goal orientation and goal consistency, and through the setting of production efficiency, quality control and cost control indicators, can quantitatively evaluate the execution effect of the production plan, set specific goals according to the actual situation of each production line, and fully consider the differences in production capacity, technical level and resource allocation, so that the goals are more practical and feasible. By formulating quality control measures, cost control strategies and delivery schedules for each production line, the production plan is refined to facilitate execution and management. By analyzing and optimizing the cost structure of the production line, cost-saving opportunities are identified to improve cost-effectiveness, and specific quality control measures are implemented to ensure that products meet quality standards and customer needs, reduce rework rates and customer complaints, and ensure on-time delivery of products through the formulation of delivery schedules and supply chain collaboration, thereby improving customer satisfaction and market competitiveness.
[0088] 2. By simulating the operation of each production line, the present invention enables enterprises to evaluate the potential impact of various plans before actual implementation, thereby improving the accuracy and effectiveness of decision-making. The use of advanced data grouping algorithms can more accurately analyze and understand production data, helping to identify patterns and trends in the production process. Through cluster analysis, enterprises can identify different modes in the operation of the production line, including efficient modes and problematic modes, so as to adjust and optimize the production process in a targeted manner. Identifying abnormal production modes enables enterprises to promptly discover and solve problems in the production process and avoid potential losses. The actual indicator values calculated based on the cluster analysis results provide an accurate and objective data basis for management to evaluate the performance of the production line and make well-founded decisions. Through cluster analysis, enterprises can clearly identify the advantages and disadvantages of each production line, which is crucial for resource allocation, process improvement and strategy adjustment. Advanced optimization methods such as cuckoo search algorithm can be used to continuously iterate and improve cluster analysis to ensure that the performance indicators of the production line are continuously improved over time. The data-driven nature of the entire process ensures the objectivity and effectiveness of decision-making and helps to achieve continuous improvement and optimization of the production process.
[0089] 3. By collecting the operating data of the production line in real time, the present invention enables the enterprise to continuously monitor the production process, discover problems in time and make adjustments. By using the reinforcement learning algorithm, the system can automatically learn and adjust the strategy to optimize the performance indicators of the production line and achieve adaptive adjustment. Based on the reinforcement learning model, the enterprise can dynamically adjust the operating parameters of the production line to improve production efficiency, reduce costs and improve product quality. By continuously monitoring the performance indicators of the production line, the effectiveness of the strategy can be evaluated to ensure continuous optimization and performance improvement of the production line.
[0090] 4. Through comparative analysis, the present invention enables enterprises to accurately evaluate the execution effect of production plans and identify areas of insufficient performance. Based on the analysis of a large amount of real-time data and reinforcement learning algorithms, the production plan can be adjusted more objectively and accurately. Through continuous comparative analysis and adjustment, enterprises can continuously improve production plans, improve production efficiency and product quality. By continuously evaluating and adjusting production plans, enterprises can more flexibly respond to changes in market demand and raw material supply. The application of real-time monitoring and reinforcement learning algorithms makes production plan adjustments more rapid, thereby improving the production efficiency of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS
[0091] The above characteristics, features and advantages of the present invention and their implementation methods and methods will become more clearly understood in conjunction with the following description of the embodiments, which are described in detail in conjunction with the accompanying drawings. Herein, a schematic diagram is shown:
[0092] Figure 1 is a flow chart of a production plan evaluation method for aviation manufacturing industry according to an embodiment of the present invention;
[0093] Figure 2 The present invention is a block diagram of a production plan evaluation system for aviation manufacturing industry according to an embodiment of the present invention.
[0094] In the figure:
[0095] 1. Goal setting and evaluation index construction module; 2. Production line scheme design module; 3. Simulation and data processing module; 4. Dynamic monitoring module; 5. Evaluation and plan adjustment module. DETAILED DESCRIPTION
[0096] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0097] According to an embodiment of the present invention, a production plan evaluation method and system for aviation manufacturing industry are provided.
[0098] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a production plan evaluation method for aviation manufacturing industry is provided, and the production plan evaluation method comprises the following steps:
[0099] S1. Establish the overall goal of the aviation manufacturing industry production plan and construct evaluation indicators that reflect the effectiveness of the production plan execution.
[0100] Preferably, the overall objectives include quality objectives, cost objectives, and delivery objectives;
[0101] Evaluation indicators include production efficiency indicators, quality control indicators and cost control indicators.
[0102] It needs to be explained that quality objectives are to determine the standards and requirements for product quality, such as the accuracy and reliability of parts and components, which involve industry standards, customer needs and internal quality control indicators of the airline.
[0103] Cost target: Set budget constraints and cost optimization targets for the entire production process, including direct costs (such as materials, labor) and indirect costs (such as equipment depreciation, management expenses).
[0104] Delivery target: According to market demand and order requirements, set the time frame of production plan to ensure production efficiency and timely delivery.
[0105] Production efficiency indicators: measure the efficiency of resource utilization and the speed of output in the production process; for example, the number of products produced per unit time, the proportion of operating time of the production line, etc.
[0106] Quality control indicators: reflect product quality level and the effectiveness of quality management; including qualification rate, return rate, customer complaint rate, etc.
[0107] Cost control indicators: evaluate the control of production costs, such as unit product cost, total production cost and budget comparison, etc.
[0108] Organize cross-departmental meetings, gather opinions and data from departments such as marketing, R&D, production, quality and finance, jointly discuss and determine overall goals, collect historical data and industry standards, conduct data-supported analysis of feasible goals for quality, cost and delivery, and based on the overall goals, define in detail the calculation method, data source and target value of each evaluation indicator, document the overall goals and evaluation indicators, and ensure through internal communication that all relevant personnel are clear about the goals and evaluation criteria.
[0109] S2. Design a specific implementation plan for each production line based on the characteristics and overall goals of the production line.
[0110] Preferably, according to the characteristics and overall objectives of the production line, designing a specific implementation plan for each production line includes the following steps:
[0111] S21. Set specific quality goals, cost goals and delivery goals for each production line based on the quality goals, cost goals and delivery goals in the overall goals and the characteristics of each production line's current production capacity, technical capabilities, staffing and resource status (conduct production line evaluation, including production data analysis, technical capability evaluation and resource research, to ensure that the goals set for each production line are consistent with the overall goals and take into account the specific conditions of the production line, so that they are achievable and operational);
[0112] S22. According to the quality objectives set for each production line, formulate quality control measures including quality inspection process, quality control points and quality improvement plan (introducing new quality inspection technology, training quality management personnel or optimizing existing quality control process);
[0113] S23. Analyze the cost structure of the production line, identify the raw material cost and energy consumption, and optimize the supply chain according to the cost target set for each production line (cost analysis and optimization are the key to improving efficiency and reducing costs; through detailed cost analysis, identify cost saving opportunities and take corresponding measures, such as reducing raw material costs through bulk purchases, or improving production processes to reduce energy consumption);
[0114] S24. Develop a delivery schedule based on the delivery targets set for each production line, ensure that the production plan matches demand and market supply, implement supply chain collaboration, and optimize management processes (the development of a delivery schedule must take into account production capacity limitations and changes in market demand, and must be carefully planned and adjusted. Supply chain collaboration and optimization of management processes are the key to ensuring production efficiency and delivery capabilities, and production plans can be optimized by improving information flow and resource allocation).
[0115] It needs to be explained that a specific implementation plan is designed for each production line based on its characteristics and overall goals, including a comprehensive assessment of each production line, taking into account its production capacity, technical capabilities, staffing and resource status, in order to set appropriate quality, cost and delivery targets; then formulating detailed quality control measures, including optimizing quality inspection processes and training quality management personnel; in-depth analysis of cost structures to identify and realize cost savings opportunities; and formulating delivery schedules and optimizing supply chain management processes to ensure efficient execution of production plans and satisfaction of market demand; ensuring that each production line can achieve its set goals efficiently and economically while responding to market and internal operational needs.
[0116] S3. Simulate the implementation plan of each production line, process the data of each production line by optimizing the data grouping algorithm, and calculate the actual index value of each production line evaluation index.
[0117] Preferably, simulating each production line implementation plan, and processing the data of each production line by optimizing the data grouping algorithm, and calculating the actual index value of each production line evaluation index includes the following steps:
[0118] S31. For each production line, collect production data on production efficiency indicators, quality control indicators and cost control indicators.
[0119] S32. Based on the production data of each production line, simulation is performed using the implementation plan of the set quality target, cost target and delivery target, and simulated production line data is generated.
[0120] The specific implementation steps are as follows: collect the current production data of each production line, including but not limited to production efficiency, quality control records, cost records and delivery time records, reconfirm the quality goals, cost goals and delivery goals set for each production line, and ensure that the simulation goals are clear and feasible; use SIMUL8 (discrete event simulation software designed for rapid modeling) in the existing technology, input the actual production data and set goals of each production line into SIMUL8, including quality control processes, cost control measures and scheduled delivery schedules, start the simulation software, and simulate the production process according to the input data and target settings; including production line operation simulation, quality inspection process simulation, cost calculation simulation and delivery time prediction, collect the production line data generated after the simulation is completed, especially the information about production efficiency, product quality, cost and expected delivery time, analyze the simulation results, especially the performance of the production line in the simulation compared with the set goals.
[0121] S33. Perform cluster analysis on the simulated production line data using an optimized data grouping algorithm.
[0122] Preferably, clustering analysis of the simulated production line data using an optimized data grouping algorithm comprises the following steps:
[0123] S331. Based on the collected production line data, select the features for clustering including production efficiency, quality control and cost control indicators (data analysis and feature selection are key, using statistical analysis and expertise to determine the indicators that best represent the performance and efficiency of the production line in order to conduct effective cluster analysis);
[0124] S332, initializing the parameters of the optimization data grouping algorithm, generating an initial population, wherein each individual in the initial population represents a parameter set of the Gaussian mixture model (initialization and population generation are the starting point of the optimization process, ensuring that there is a population with sufficient diversity to cover the potential solution space so that the algorithm can find the optimal or near-optimal clustering solution);
[0125] S333. Use each individual in the initial population (i.e., each set of Gaussian mixture model parameter sets) to cluster the production line data (clustering attempt is a key step in the experimental process, and it is necessary to run the algorithm to perform multiple clustering analyses on the data to find the clustering model that can most accurately reflect the performance characteristics of the production line);
[0126] S334. Use the maximum likelihood function to calculate the fitness of each individual in the initial population (clustering attempts are a key step in the experimental process, and it is necessary to run the algorithm to perform multiple clustering analyses on the data to find the clustering model that most accurately reflects the performance characteristics of the production line);
[0127] Among them, the formula for calculating the fitness of each individual in the initial population using the maximum likelihood function is:
[0128] ;
[0129] In the formula, Represents the fitness of the parameter set in the Gaussian mixture model;
[0130] n represents the total number of data points in the parameter set;
[0131] Indicates the first i data points;
[0132] K represents the number of Gaussian distributions;
[0133] Indicates k A mixture of Gaussian distributions with weights;
[0134] Indicates k The mean vector of the Gaussian distribution;
[0135] Indicates k The covariance matrix of the Gaussian distribution;
[0136] represents the probability density function of the Gaussian distribution;
[0137] i means Data points in the parameter set;
[0138] k represents Gaussian distribution.
[0139] It should be noted that the optimized data grouping algorithm is GMM-CS. GMM-CS (Gaussian MixtureModel-Cuckoo Search) is an optimized data grouping algorithm that combines the Gaussian mixture model (GMM) and the cuckoo search (CS) algorithm. It improves the effect of data clustering through optimization technology, especially when dealing with complex or high-dimensional data sets.
[0140] GMM-CS combines the clustering ability of GMM and the optimization ability of CS algorithm to find the best clustering parameter settings, that is, to determine the Gaussian distribution parameters that best represent the data structure; specifically, GMM-CS is performed through the following steps:
[0141] Initialization: Initialize the population of Gaussian mixture model parameters based on the dataset (i.e. the bird's nest in the cuckoo algorithm).
[0142] Cuckoo search optimization: Use the cuckoo search algorithm to explore the parameter space and find the optimal Gaussian mixture model parameters. This includes generating new solutions (new sets of Gaussian mixture model parameters) and selecting the best solution by comparing their fitness (e.g., using maximum likelihood estimation).
[0143] Fitness evaluation: Use the maximum likelihood function or other evaluation criteria to calculate the fitness of each Gaussian mixture model parameter set to evaluate the clustering effect.
[0144] Iterative update: Update the solution set according to the cuckoo search algorithm, and continuously iterate to find better Gaussian mixture model parameter settings.
[0145] Convergence judgment: When the maximum number of iterations is reached or the fitness meets a certain predetermined condition, the algorithm stops and outputs the optimal Gaussian mixture model parameter set.
[0146] The Gaussian mixture model is a probabilistic model used to represent a data set containing multiple sub-distributions. The sub-distributions are assumed to be Gaussian distributions (also called normal distributions). Each sub-distribution corresponds to a "cluster" or group in the data. In the Gaussian mixture model, the Gaussian distribution of each cluster is determined by two parameters: the mean and variance, which are the mean vector and the covariance matrix in the multidimensional case. The key feature of the model is that it not only provides clustering functionality, but also estimates the probability that each data point belongs to each cluster, that is, it is a "soft" clustering method.
[0147] Maximum likelihood estimation is a statistical method used to estimate model parameters based on observed data. Its basic idea is that given a set of data and a probability model, maximum likelihood estimation will find a set of parameter values that maximizes the probability (i.e., likelihood) of this set of data appearing under the parameters.
[0148] In the maximum likelihood function, likelihood is defined as the probability of the observed data appearing under given parameters. For Gaussian mixture models, the goal of maximum likelihood estimation is to find a set of parameters that maximizes the product of the probabilities of all data points appearing in the model. Since directly maximizing the likelihood function is usually more complicated, it is often simplified by maximizing its logarithmic form, which involves mathematical iterative optimization algorithms, such as the EM algorithm, which alternately performs two steps: in the expectation (E) step, the probability of the data belonging to each component is calculated (i.e., "soft" clustering); in the maximization (M) step, the model parameters are updated to maximize the likelihood function.
[0149] S34. Use the cuckoo search algorithm to iteratively optimize the production line data after cluster analysis, and finally select the clustering result with the highest fitness.
[0150] Preferably, using the cuckoo search algorithm to iteratively optimize the production line data after cluster analysis and finally selecting the cluster result with the highest fitness includes the following steps:
[0151] S341. Randomly select individuals in the population as cuckoos and generate new solutions through Levy flight (Levi flight is the key feature of the cuckoo search algorithm, which allows the algorithm to explore long distances, helps to jump out of the local optimal solution, and increases the possibility of finding the global optimal solution);
[0152] S342. Calculate the fitness of the newly generated solution using the maximum likelihood function (fitness calculation is the basis for decision-making in the optimization process, and solutions with high fitness are more likely to be retained and used for further iterations);
[0153] S343. If the fitness of the newly generated solution is greater than the fitness of the cuckoo individual selected as its parent, or if it is randomly selected for replacement based on the discovery probability (a parameter in the algorithm that controls the balance between exploration and exploitation), the current solution is replaced with the new solution (the key to the algorithm optimization process, by continuously replacing solutions with lower fitness, the population gradually evolves towards a better solution);
[0154] S344. In each iteration, the optimal solution encountered is recorded and updated, and the optimal solution recorded during the iteration is used as the final clustering model parameters (continuously recording and updating the optimal solution ensures that the algorithm can remember and ultimately provide the best clustering parameters found during the iteration).
[0155] In addition, it should be noted that the cuckoo search algorithm is used to iteratively optimize the production line data after clustering analysis to improve the accuracy and efficiency of the clustering results. First, new solutions are generated through random selection and Levy flight to explore the solution space and try to jump out of the local optimum. Next, the fitness of the new solution is calculated using the maximum likelihood function, and the decision on whether to replace the current solution is based on the fitness evaluation, which helps to continuously evolve to a better solution. In each iteration, the optimal solution encountered is recorded and updated, and finally the clustering model parameters with the highest fitness are selected, which can effectively optimize the clustering analysis and find the clustering results that best reflect the actual operation mode of the production line.
[0156] S35. Use the clustering results to identify abnormal patterns in production line operation and analyze the strengths and weaknesses of each production line.
[0157] First, carefully analyze the optimized clustering results, including a detailed review of each cluster (i.e., production line operation mode) to identify commonalities and differences. The analysis of clustering results is the starting point of this stage. Visualization tools (such as scatter plots, heat maps, etc.) and statistical analysis methods are needed to explore the differences and connections between clusters, especially those clusters that differ significantly from the average performance.
[0158] Based on cluster analysis, production line operation patterns with abnormally low production efficiency or abnormally high costs are identified; operation patterns indicate problems or bottlenecks in the production process. The identification of abnormal patterns relies on a deep understanding of the clustering results and expertise in the production process. It is necessary to combine data-driven insights with domain knowledge to accurately identify which patterns are abnormal.
[0159] For each cluster (representing one or more production line operation modes), analyze its performance indicators (such as production efficiency, quality control and cost-effectiveness) to determine the advantages and disadvantages. The advantages and disadvantages analysis of each production line requires a detailed examination of its performance on different indicators. By comparing the performance differences between different clusters (or modes), we can identify practices that help improve efficiency or reduce costs, and those practices that are the causes of poor performance.
[0160] Based on the identified abnormal patterns and the advantages and disadvantages obtained through analysis, targeted improvement measures are proposed, including optimizing production processes, adjusting resource allocation, improving quality control processes, etc.
[0161] S36. Based on the results of cluster analysis, the actual index values of each production line in terms of production efficiency index, quality control index and cost control index are calculated.
[0162] Preferably, based on the results of cluster analysis, calculating the actual index values of each production line in terms of production efficiency index, quality control index and cost control index comprises the following steps:
[0163] S361, obtaining the results of cluster analysis, and classifying each production line into a corresponding cluster category (first, obtaining the cluster results of each production line from the previous cluster analysis, that is, determining which cluster category each production line belongs to, which is usually performed automatically after the cluster analysis is completed, and each production line is marked as belonging to a specific cluster according to its characteristics);
[0164] S362. For each cluster, calculate the average values of production efficiency indicators, quality control indicators and cost control indicators for all production lines within each cluster category (the average value calculation provides a comprehensive view of the performance of production lines within a cluster category. By averaging the performance indicators within the cluster, the overall performance of the production lines in this category can be better understood);
[0165] S363, mapping the average value in the clustering result to each production line as the actual index value of each production line on the corresponding index (the obtained average index value of each cluster category is mapped to each production line in the cluster as the actual index value of these production lines on the corresponding index, and the production lines belonging to the same cluster will share the same average value on these indexes);
[0166] The average calculation formula for production efficiency index, quality control index and cost control index is:
[0167] ;
[0168] ;
[0169] ;
[0170] In the formula, Indicates j The number of production lines in a cluster;
[0171] Indicates j Cluster categories;
[0172] Indicates j The average production efficiency index of each cluster category;
[0173] It represents the production efficiency index;
[0174] Indicates j The average quality control index of each cluster category;
[0175] Indicates quality control indicators;
[0176] Indicates jThe average cost control index of each cluster category;
[0177] Represents cost control indicators.
[0178] In addition, it should be noted that based on the results of cluster analysis, the actual performance of each production line in terms of key performance indicators is calculated, which not only provides data support for understanding and evaluating the performance of each production line, but also lays the foundation for formulating targeted improvement measures.
[0179] S4. Establish a dynamic monitoring mechanism and use reinforcement learning algorithms to analyze the data of each production line.
[0180] It should be explained that, first of all, a data collection system needs to be established, which can collect the operation data of each production line in real time, including but not limited to production efficiency, quality control data, cost data, etc.;
[0181] Design a reinforcement learning model and automatically learn and adjust strategies based on the collected data to optimize the performance indicators of the production line; define the state space (such as various operating parameters of the production line), action space (measures that can be taken), and reward function (evaluate the quality of actions based on performance indicators).
[0182] The reinforcement learning model is trained by interacting with actual data from the production line. During this process, the model will continuously try different actions (i.e. adjust the strategy) and learn and optimize these strategies based on the feedback of the reward function.
[0183] Based on the trained reinforcement learning model, an optimization strategy is implemented to dynamically adjust the operating parameters of the production line. At the same time, the performance indicators of the production line are continuously monitored to evaluate the effectiveness of the strategy.
[0184] The task of designing a reinforcement learning model to optimize the performance indicators of a production line requires the detailed definition of the three core components of the model: state space, action space, and reward function. The following is a concrete implementation of this task:
[0185] The state space defines all possible environmental states, that is, all the information the model has at any given point in time. For a production line, the state can include but is not limited to: production rate (output per unit time), current production line failure rate, number of work-in-progress (WIP), equipment utilization, quality control indicators such as defect rate and action space (A).
[0186] The action space defines all possible actions that the model can take. These actions are specific measures that the model can take when trying to optimize the performance of the production line, such as: adjusting the production speed, replacing or repairing equipment, adjusting the staffing on the line, changing the raw material supplier, implementing quality improvement measures, and the reward function (R).
[0187] The reward function is a criterion for evaluating the change in the state of the environment after taking a certain action. It is usually set according to the performance indicators of the production line. The reward function needs to clearly reflect the goal, such as improving production efficiency, reducing costs, and improving product quality. The reward function can be designed according to the following criteria:
[0188] If the action leads to increased production efficiency, the reward value increases positively.
[0189] If the action leads to cost reduction (such as reducing the waste of raw materials and reducing energy consumption), the reward value increases positively.
[0190] If the action improves product quality (such as reducing the defect rate), the reward value increases positively.
[0191] If the action leads to any negative impact (such as increased failure rate, reduced production speed), the reward value increases negatively.
[0192] S5. Compare the target value in the evaluation index with the actual index value to evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry based on the analysis results of the reinforcement learning algorithm.
[0193] Preferably, comparing the target value in the evaluation index with the actual index value, evaluating the overall effect of the production plan, and adjusting the production plan of the aviation manufacturing industry in combination with the analysis results of the reinforcement learning algorithm include the following steps:
[0194] S51. Collect the actual operation data of each production line and the actual index values of the evaluation indicators (it is crucial to ensure the accuracy and completeness of the data, which will directly affect the quality and accuracy of subsequent analysis);
[0195] S52. Match the actual index value with the target value set in the production plan and compare them (through comparative analysis, you can intuitively identify which indicators have reached the target and which have not, providing a basis for analysis);
[0196] S53. Evaluate the overall execution effect of the production plan based on the comparison results (in-depth analysis of the performance of each indicator and the reasons for the differences is the key to evaluating the execution effect of the production plan, which requires not only data analysis but also an understanding of the actual production process);
[0197] S54. Obtain the results of reinforcement learning algorithm analysis (it is crucial to understand and utilize the analysis results provided by reinforcement learning algorithms, as the results are generated based on large amounts of data and complex algorithms and can provide in-depth insights and recommendations);
[0198] S55. Based on the overall execution effect of the evaluated production plan and the results of the reinforcement learning algorithm analysis, identify the problems in the execution of the production plan and adjust the production plan of the aviation manufacturing industry (the key to converting all the previous analysis work into practical actions. Adjusting the production plan requires comprehensive consideration of various information and suggestions to ensure that the optimization of the production plan is comprehensive and practical).
[0199] Preferably, based on the comparison results, evaluating the overall execution effect of the production plan includes the following steps:
[0200] S531, analyzing the difference between the actual indicator value of each evaluation indicator and the preset target value;
[0201] S532. Based on the analysis results of the differences, identify the performance deficiencies in the production plan execution;
[0202] S533. For each aspect of insufficient performance, analyze the reasons why the actual indicators did not reach the target value;
[0203] S534. Integrate the actual performance of all evaluation indicators, evaluate the overall execution effect of the production plan, and summarize the overall performance of the production plan execution.
[0204] It needs to be explained that the differences between the actual and target values of the evaluation indicators are analyzed, the performance deficiencies in the execution of the production plan are identified and analyzed, and the reasons for the failure to meet the indicators are explored in depth to comprehensively evaluate the overall execution effect of the production plan; this process not only reveals specific problems in production activities, but also provides guidance on how to adjust and optimize production plans based on these findings; by integrating data analysis, cross-departmental collaboration and expertise, an overall evaluation of production execution is formed, thereby clarifying production strengths and weaknesses, and improving measures are proposed accordingly. This evaluation and optimization cycle is the key to continuously improving production efficiency and product quality.
[0205] According to another aspect of the present invention, Figure 2 As shown, a production plan evaluation system for aviation manufacturing industry is also provided, the system comprising: a target setting and evaluation index construction module 1, a production line scheme design module 2, a simulation and data processing module 3, a dynamic monitoring module 4 and an evaluation and plan adjustment module 5;
[0206] The target setting and evaluation index construction module 1 is connected to the simulation and data processing module 3 through the production line scheme design module 2, and the simulation and data processing module 3 is connected to the evaluation and plan adjustment module 5 through the dynamic monitoring module 4;
[0207] Goal setting and evaluation index construction module 1 is used to formulate the overall goal of the aviation manufacturing production plan and construct evaluation indicators that reflect the effect of production plan execution;
[0208] Production line solution design module 2 is used to design a specific implementation plan for each production line based on the characteristics and overall goals of the production line;
[0209] The simulation and data processing module 3 is used to simulate the implementation plan of each production line, process the data of each production line by optimizing the data grouping algorithm, and calculate the actual index value of the evaluation index of each production line;
[0210] Dynamic monitoring module 4, used to establish a dynamic monitoring mechanism and analyze the data of each production line using a reinforcement learning algorithm;
[0211] The evaluation and plan adjustment module 5 is used to compare the target value in the evaluation index with the actual index value, evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry in combination with the analysis results of the reinforcement learning algorithm.
[0212] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention sets quality, cost and delivery targets to provide clear guidance for production activities, help various departments focus on common goals, enhance goal orientation and goal consistency, and through the setting of production efficiency, quality control and cost control indicators, the execution effect of the production plan can be quantitatively evaluated, and specific goals can be set according to the actual situation of each production line, fully considering the differences in production capacity, technical level and resource allocation, so that the goals are more practical and feasible. By formulating quality control measures, cost control strategies and delivery schedules for each production line, the production plan is refined to facilitate execution and management. By analyzing and optimizing the cost structure of the production line, cost-saving opportunities are identified, cost-effectiveness is improved, and specific quality control measures are implemented to ensure that products meet quality standards and customer needs, reduce rework rates and customer complaints, and ensure on-time delivery of products through the formulation of delivery schedules and supply chain collaboration, thereby improving customer satisfaction and market competitiveness. The present invention simulates the operation of each production line, so that enterprises can evaluate before actual implementation. The potential impact of various plans can improve the accuracy and effectiveness of decision-making. Advanced data grouping algorithms can more accurately analyze and understand production data and help identify patterns and trends in the production process. Through cluster analysis, enterprises can identify different modes in the operation of production lines, including efficient modes and problematic modes, so as to adjust and optimize production processes in a targeted manner. Identifying abnormal production modes enables enterprises to promptly discover and solve problems in the production process and avoid potential losses. The actual indicator values calculated based on the cluster analysis results provide management with an accurate and objective data basis for evaluating the performance of production lines and making well-founded decisions. Through cluster analysis, enterprises can clearly identify the advantages and disadvantages of each production line, which is crucial for resource allocation, process improvement and strategy adjustment. Advanced optimization methods such as cuckoo search algorithms can continuously iterate and improve cluster analysis to ensure that the performance indicators of production lines are continuously improved over time. The data-driven nature of the entire process ensures the objectivity and effectiveness of decision-making and helps to achieve continuous improvement and optimization of the production process.The present invention collects the operation data of the production line in real time, so that the enterprise can continuously monitor the production process, find problems in time and make adjustments. By using the reinforcement learning algorithm, the system can automatically learn and adjust the strategy to optimize the performance indicators of the production line and realize adaptive adjustment. Based on the reinforcement learning model, the enterprise can dynamically adjust the operation parameters of the production line to improve production efficiency, reduce costs and improve product quality. The continuous monitoring of the performance indicators of the production line can evaluate the effect of the strategy to ensure the continuous optimization and performance improvement of the production line. Through comparative analysis, the present invention can accurately evaluate the execution effect of the production plan and identify the aspects of insufficient performance. Based on the analysis of a large amount of real-time data and reinforcement learning algorithms, the adjustment of the production plan is more objective and accurate. Through continuous comparative analysis and adjustment, the enterprise can continuously improve the production plan, improve production efficiency and product quality. Through continuous evaluation and adjustment of the production plan, the enterprise can more flexibly respond to changes in market demand and raw material supply. The application of real-time monitoring and reinforcement learning algorithms makes the adjustment of production plans more rapid, improving the production efficiency of the enterprise. ;
[0213] Although the present invention has been disclosed as above with preferred embodiments, the embodiments are merely examples for the convenience of description and are not intended to limit the present invention. Those skilled in the art may make several changes and modifications without departing from the spirit and scope of the present invention. The scope of protection claimed by the present invention shall be based on the claims.
Claims
1. A production plan evaluation method for aviation manufacturing industry, characterized in that: The production plan evaluation method comprises the following steps: S1. Formulate the overall goal of the aviation manufacturing industry production plan and construct evaluation indicators that reflect the effectiveness of the production plan execution; S2. Design a specific implementation plan for each production line based on its characteristics and overall goals; S3, simulating the implementation plan of each production line, and processing the data of each production line by optimizing the data grouping algorithm, and calculating the actual index value of the evaluation index of each production line; S4. Establish a dynamic monitoring mechanism and use reinforcement learning algorithms to analyze the data of each production line; S5. Compare the target value in the evaluation index with the actual index value to evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry in combination with the analysis results of the reinforcement learning algorithm; The method of simulating the implementation plan of each production line, processing the data of each production line by optimizing the data grouping algorithm, and calculating the actual index value of the evaluation index of each production line includes the following steps: S31. For each production line, collect production data on production efficiency indicators, quality control indicators and cost control indicators; S32. Based on the production data of each production line, simulation is performed using the implementation plan of the set quality target, cost target and delivery target, and simulated production line data is generated; S33, using an optimized data grouping algorithm to perform cluster analysis on the simulated production line data; S34, using the cuckoo search algorithm to iteratively optimize the production line data after cluster analysis, and finally select the clustering result with the highest fitness; S35, using the clustering results to identify abnormal patterns in the operation of the production line and analyze the advantages and disadvantages of each production line; S36. Based on the results of the cluster analysis, calculate the actual index values of each production line in terms of production efficiency index, quality control index and cost control index; The cluster analysis of the simulated production line data using the optimized data grouping algorithm comprises the following steps: S331. Based on the collected production line data, select clustered features including production efficiency, quality control and cost control indicators; S332, initializing the parameters of the optimized data grouping algorithm to generate an initial population, wherein each individual in the initial population represents a parameter set of a Gaussian mixture model; S333, using each individual in the initial population to attempt to cluster the production line data; S334, using the maximum likelihood function to calculate the fitness of each individual in the initial population; The formula for calculating the fitness of each individual in the initial population using the maximum likelihood function is: ; In the formula, Represents the fitness of the parameter set in the Gaussian mixture model; n represents the total number of data points in the parameter set; Indicates the first i data points; K represents the number of Gaussian distributions; Indicates k The mixture weights of the Gaussian distribution; Indicates k The mean vector of the Gaussian distribution; Indicates k The covariance matrix of the Gaussian distribution; represents the probability density function of the Gaussian distribution; i means Data points in the parameter set; k represents Gaussian distribution.
2. A production plan evaluation method for aviation manufacturing industry according to claim 1, characterized in that: The overall objectives include quality objectives, cost objectives and delivery objectives; The evaluation indicators include production efficiency indicators, quality control indicators and cost control indicators.
3. A production plan evaluation method for aviation manufacturing industry according to claim 1, characterized in that: According to the characteristics and overall objectives of the production line, the specific implementation plan for each production line includes the following steps: S21. Set specific quality goals, cost goals and delivery goals for each production line based on the quality goals, cost goals and delivery goals in the overall goals and the characteristics of each production line's current production capacity, technical capabilities, staffing and resource status; S22. Develop quality control measures including quality inspection process, quality control points and quality improvement plan according to the quality objectives set for each production line; S23. Analyze the cost structure of each production line, identify raw material costs and energy consumption, and optimize the supply chain according to the cost targets set for each production line; S24. Develop a delivery schedule based on the delivery targets set for each production line, ensure that the production plan matches demand and market supply, implement supply chain collaboration, and optimize management processes.
4. A production plan evaluation method for aviation manufacturing industry according to claim 1, characterized in that: The method of iteratively optimizing the production line data after cluster analysis by using the cuckoo search algorithm and finally selecting the cluster result with the highest fitness includes the following steps: S341, randomly select individuals in the population as cuckoos, and generate new solutions through Levy flight; S342, using the maximum likelihood function to calculate the fitness of the newly generated solution; S343. If the fitness of the newly generated solution is greater than the fitness of the cuckoo individual selected as the parent, or if it is randomly selected for replacement according to the discovery probability, the current solution is replaced by the new solution; S344. In each iteration, the optimal solution encountered is recorded and updated, and the optimal solution recorded during the iteration process is used as the final clustering model parameter.
5. A production plan evaluation method for aviation manufacturing industry according to claim 4, characterized in that: The calculation of the actual index values of each production line on the production efficiency index, quality control index and cost control index based on the cluster analysis results includes the following steps: S361, obtaining the result of cluster analysis, and classifying each production line into a corresponding cluster category; S362. For each cluster, calculate the average values of production efficiency index, quality control index and cost control index of all production lines in each cluster category; S363, mapping the average value in the clustering result to each production line as the actual index value of each production line on the corresponding index; The calculation formula for the average value of the production efficiency index, quality control index and cost control index is: ; ; ; In the formula, Indicates j The number of production lines in a cluster; Indicates j Cluster categories; Indicates j The average production efficiency index of each cluster category; It represents the production efficiency index; Indicates j The average quality control index of each cluster category; Indicates quality control indicators; Indicates j The average cost control index of each cluster category; Represents cost control indicators.
6. A production plan evaluation method for aviation manufacturing industry according to claim 5, characterized in that: The above-mentioned comparing the target value in the evaluation index with the actual index value, evaluating the overall effect of the production plan, and adjusting the production plan of the aviation manufacturing industry in combination with the analysis results of the reinforcement learning algorithm includes the following steps: S51, collecting actual operation data of each production line and actual index values of evaluation indexes; S52, matching the actual index value with the target value set in the production plan and comparing them; S53. Evaluate the overall execution effect of the production plan based on the comparison results; S54, obtaining the result of reinforcement learning algorithm analysis; S55. Based on the overall execution effect of the evaluated production plan and the results of the reinforcement learning algorithm analysis, identify the problems in the execution of the production plan and adjust the production plan of the aviation manufacturing industry.
7. A production plan evaluation method for aviation manufacturing industry according to claim 6, characterized in that: According to the comparison results, evaluating the overall execution effect of the production plan includes the following steps: S531, analyzing the difference between the actual indicator value of each evaluation indicator and the preset target value; S532. Based on the analysis results of the differences, identify the performance deficiencies in the production plan execution; S533. For each aspect of insufficient performance, analyze the reasons why the actual indicators did not reach the target value; S534. Integrate the actual performance of all evaluation indicators, evaluate the overall execution effect of the production plan, and summarize the overall performance of the production plan execution.
8. A production plan evaluation system for aviation manufacturing industry, used to implement the production plan evaluation method for aviation manufacturing industry according to any one of claims 1 to 7, characterized in that: The system includes: target setting and evaluation index construction module, production line scheme design module, simulation and data processing module, dynamic monitoring module and evaluation and plan adjustment module; The target setting and evaluation index construction module is connected with the simulation and data processing module through the production line scheme design module, and the simulation and data processing module is connected with the evaluation and plan adjustment module through the dynamic monitoring module; The module for goal setting and evaluation index construction is used to formulate the overall goal of the aviation manufacturing production plan and construct evaluation indicators that reflect the effect of production plan execution; Production line solution design module, used to design specific implementation plans for each production line based on the characteristics and overall goals of the production line; The simulation and data processing module is used to simulate the implementation plan of each production line, process the data of each production line by optimizing the data grouping algorithm, and calculate the actual index value of the evaluation index of each production line; Dynamic monitoring module, which is used to establish a dynamic monitoring mechanism and analyze the data of each production line using reinforcement learning algorithms; The evaluation and plan adjustment module is used to compare the target value in the evaluation index with the actual index value, evaluate the overall effect of the production plan, and adjust the production plan of the aviation manufacturing industry based on the analysis results of the reinforcement learning algorithm.
