Cost optimization algorithm module for manufacturing execution system MES

Abstract

The invention relates to a cost optimization algorithm module for an MES (Manufacturing Execution System). The cost optimization algorithm module comprises a data acquisition unit, a data preprocessing unit, a cost optimization model construction unit and an optimization algorithm unit, the data acquisition unit is used for acquiring multi-dimensional data in real time; the data preprocessing unit is used for preprocessing the multi-dimensional data acquired by the data acquisition unit; the cost optimization model construction unit constructs a cost optimization model based on a multi-objective optimization method according to the data preprocessed by the data and processing unit; the optimization algorithm unit combines a genetic algorithm and a simulated annealing algorithm, adopts an intelligent search strategy to solve the model, and generates an optimal production plan and resource allocation scheme; dynamically updating an optimization result to a manufacturing execution system MES, and performing dynamic adjustment and feedback; according to the method, multiple dimensions such as production planning, equipment utilization, inventory management and quality control are comprehensively considered, and cost optimization is carried out on each link in the production process.

Description

Technical Field

[0001] The present invention relates to the technical field of manufacturing execution system MES, and particularly to a cost optimization algorithm module for manufacturing execution system MES. Background Art

[0002] Manufacturing Execution System MES is a core information management system in modern manufacturing, which is used to connect the production site and the enterprise management system, providing functions such as production scheduling, data collection, equipment monitoring, inventory management, etc. However, in actual applications, existing MES systems often struggle to cope with the increasingly complex production processes and changing market demands in the manufacturing industry. Especially in cost management, existing solutions often fail to achieve refined control over costs in multiple links and dimensions.

[0003] Traditional cost control mainly relies on linear programming or manual experience judgment, lacking the ability to process large-scale and multi-dimensional data, and unable to meet the efficient and precise requirements of manufacturing enterprises for cost optimization in the era of intelligent manufacturing. Therefore, there is an urgent need for a cost optimization module that can integrate real-time data analysis and intelligent optimization algorithms to help enterprises minimize costs and optimize resource allocation in complex production environments.

[0004] Chinese Patent with Publication No. CN118409548A discloses a manufacturing cloud platform based on MES system, including requirements analysis technology, system integration technology, data collection technology, data processing technology, process control technology, data security technology, mobile technology, artificial intelligence technology. Through requirements analysis, the development team improves the functional requirements, operation requirements and interface requirements of the system according to customer needs to dock with customers, and coordinates internal production activities of the enterprise by means of real-time monitoring of the production process, collection of production data, optimization of production plans, management of production processes, improvement of production efficiency and production quality, etc., improving the production efficiency of the enterprise, reducing production costs and improving product quality, effectively controlling the production process, enabling traceability, and strengthening product quality management; however, this patent can only bring the technical effect of reducing production costs and cannot optimize and dynamically adjust costs. Summary of the Invention

[0005] The present invention provides a cost optimization algorithm module for manufacturing execution system MES, which can comprehensively consider multiple dimensions such as production plan, equipment utilization, inventory management and quality control, and optimize the costs of each link in the production process through real-time data collection and intelligent algorithm solving, providing an efficient and intelligent cost management tool for manufacturing enterprises.

[0006] To achieve the above object, the present invention is implemented by adopting the following technical solutions: A cost optimization algorithm module for manufacturing the Manufacturing Execution System (MES), including a data acquisition unit, a data preprocessing unit, a cost optimization model construction unit, and an optimization algorithm unit; The data acquisition unit is used to obtain multi-dimensional data related to production, equipment, inventory, and quality in real time; The data preprocessing unit is used to clean, normalize, and extract features from the multi-dimensional data obtained by the data acquisition unit; The cost optimization model construction unit constructs a cost optimization model based on the data preprocessed by the data and processing unit, using a multi-objective optimization method; The optimization algorithm unit combines the genetic algorithm and the simulated annealing algorithm, and uses an intelligent search strategy to solve the model, generating an optimal production plan and resource allocation plan; The cost optimization algorithm module dynamically updates the optimization results to the Manufacturing Execution System (MES), performs dynamic adjustment and feedback, ensures that the production plan is synchronized with the actual situation, and continuously optimizes the algorithm parameters through historical data analysis and machine learning.

[0007] Further, the multi-dimensional data includes raw material procurement costs, real-time inventory information, process parameter data during production, equipment operation status data and performance indicators, inventory management data, product quality inspection data, quality control records, customer feedback information, current production schedules, and order information. The equipment operation status data includes the equipment's startup time, shutdown time, fault records, and maintenance history. The product quality inspection data includes the inspection pass rate, rework rate, and customer feedback information for each production batch.

[0008] Further, the cost optimization model construction unit uses a multi-objective optimization method. The objective function of the model includes minimizing the total production cost, minimizing the inventory holding cost, maximizing the equipment utilization rate, and maximizing the product quality pass rate. The constraint conditions of the model include production capacity constraints, inventory capacity constraints, equipment availability constraints, and quality standard constraints.

[0009] Further, the historical data analysis includes identifying and learning the patterns and trends in the production process.

[0010] Further, the dynamic adjustment and feedback means dynamically optimizing according to the real-time changes in the production environment, real-time adjusting the production plan and resource allocation, and feeding back the optimization results to each module of the Manufacturing Execution System (MES).

[0011] Compared with the prior art, the beneficial effects of the present invention are: 1) Comprehensive cost optimization: By comprehensively optimizing production, inventory, equipment, and quality, etc., the total production cost is significantly reduced; 2) Real-time performance and dynamic adjustment: The cost optimization algorithm can collect production data in real time and make dynamic adjustments to ensure that the production plan is synchronized with the actual situation and improve the response speed; 3) Flexibility and scalability: The algorithm module can be flexibly configured to adapt to different production environments and requirements, and can be extended according to the scale and complexity of the enterprise. Specific implementation manner

[0012] The following further describes the specific implementation manner of the present invention: A cost optimization algorithm module for a manufacturing execution system MES of the present invention includes a data acquisition unit, a data preprocessing unit, a cost optimization model construction unit, and an optimization algorithm unit; 1) The data acquisition unit, through the manufacturing execution system MES interface, obtains multi-dimensional data related to production, equipment, inventory, and quality in real time. The multi-dimensional data includes: (1) The raw material procurement cost, reflecting the impact of raw material price fluctuations on the overall cost; (2) Process parameter data during the production process, such as the processing time of each process, the equipment operation load, and the human resource allocation; (3) The operation status data of the equipment, including the equipment startup time, shutdown time, fault records, and maintenance history; (4) Inventory management data, including the inventory levels of various materials, inventory turnover rates, order delivery cycles, etc.; (5) Product quality data, including the inspection pass rate, rework rate of each production batch, and information feedback by customers; The data acquisition unit realizes data acquisition through the interface with the MES system. The interfaces include: The interface with the enterprise resource planning ERP system, used to obtain the raw material procurement cost and real-time inventory information; The interface with the production equipment real-time monitoring system, used to obtain the equipment operation status and performance indicators in real time; The interface with the quality management system, used to obtain the product quality inspection data, quality control records, and customer feedback information; The interface with the production planning system, used to obtain the current production schedule and order information.

[0013] 2) The data preprocessing unit performs preprocessing operations such as data cleaning, normalization, and feature extraction on the collected multi-dimensional data to ensure the accuracy and consistency of the data. At the same time, using statistical analysis and data mining techniques, it analyzes the characteristics and trends of the data and extracts key factors related to cost optimization; The data cleaning includes identifying and removing outliers in the data, such as measurement values that deviate significantly from the normal range, and filling in missing values using interpolation, mean method or mode method to ensure data integrity; The normalization process includes converting data with different dimensions and ranges into a unified standard. Common methods include min-max normalization and Z-score standardization to improve data consistency; The feature extraction includes extracting key features from the original data, such as calculating the average load rate of the device, production efficiency indicators, average quality loss of each product, etc.; performing correlation analysis to identify variables highly correlated with production cost optimization to help the algorithm optimize more accurately.

[0014] 3) The cost optimization model construction unit constructs a cost optimization model based on the data preprocessed by the data and processing unit using a multi-objective optimization method; the objective function of the model includes minimizing the total production cost, minimizing the inventory holding cost, maximizing the equipment utilization rate and maximizing the product quality pass rate, Minimizing the total production cost includes raw material cost, labor cost, energy consumption cost and equipment maintenance cost, etc.; Minimizing the inventory holding cost takes into account the cost of occupying inventory funds, storage expenses, and the risk of obsolescence due to overstocking; Maximizing the equipment utilization rate improves the operating efficiency of the equipment and reduces idle time by reasonably scheduling production tasks; Maximizing the product quality pass rate reduces the defective rate, improves the quality control level during the production process, and ensures that the products meet customer requirements; The constraint conditions of the model include: Production capacity constraint: Ensure that the production plan is within the established production capacity range; Equipment utilization constraint: Avoid equipment overload or vacancy; Inventory capacity constraint: Prevent overstocking or understocking; Quality pass rate constraint: Ensure that the products produced meet the quality standards.

[0015] 4) The optimization algorithm unit combines the genetic algorithm and the simulated annealing algorithm, and uses an intelligent search strategy to solve the model to generate an optimal production plan and resource allocation plan; Genetic algorithm: Used to generate an initial solution space and perform global search to find a relatively good initial cost optimization solution. First, generate a set of initial solutions, and iteratively optimize the solution set through selection, crossover and mutation operations, and retain the relatively good solutions. The specific process is as follows, (1) Population initialization: Randomly generate a certain number of candidate solutions (individuals), each solution represents a possible production plan and resource allocation plan, and the diversity of the initial solutions helps the comprehensiveness of the search; (2)Fitness evaluation: Calculate the fitness value of each candidate solution by comprehensively considering indicators such as production cost, equipment utilization rate, and product quality; (3)Selection operation: Adopt roulette wheel selection or tournament selection strategy to select individuals with high fitness to enter the next generation, increasing the genetic probability of excellent individuals; (4)Crossover operation: Select two or more individuals for crossover to generate new individuals, such as using single-point crossover, two-point crossover, or uniform crossover to combine excellent features; (5)Mutation operation: Make small random changes to some individuals to increase the diversity of the population and prevent premature convergence to a local optimal solution; (6)Iterative process: Repeat the steps of fitness evaluation, selection, crossover, and mutation until a predetermined stopping condition is reached, such as stable fitness or reaching the maximum number of iterations; For the simulated annealing algorithm, using the optimal solution of the genetic algorithm as the initial solution, further optimize within the local search space to prevent falling into a local optimal solution. The specific steps are as follows: (1)Initial solution setting: Select a random initial production plan as the current solution and set a relatively high initial temperature; (2)Neighboring solution generation: Generate a neighboring solution by slightly adjusting some parameters based on the current solution; (3)Fitness comparison: Evaluate the fitness of the neighboring solution and compare its advantages and disadvantages; (4)Acceptance criterion: Determine whether to accept the new solution according to the fitness difference. Adopt the Metropolis criterion. If the neighboring solution is better than the current solution, accept it; if not, accept it according to the probability to maintain the exploratory nature; (5)Temperature reduction: Gradually reduce the temperature according to a predetermined temperature reduction strategy, usually using exponential decay or linear decay; (6)Stopping condition: When the temperature drops to a preset threshold or reaches the maximum number of iterations, stop the algorithm and output the current optimal solution.

[0016] 5)Optimization result output and application: After the optimization algorithm calculates the optimal solution, output the results to the MES system, including: Production plan adjustment suggestions: Provide the optimal production scheduling plan; Inventory management suggestions: Provide the optimal inventory ordering and replenishment plan; Equipment scheduling suggestions: Optimize the operation and maintenance plan of the equipment to reduce the equipment operation cost; Quality control plan: Provide improvement suggestions for quality management to reduce the defective product rate.

[0017] 6)The cost optimization algorithm module has the ability of self-learning and self-adaptation, including: (1) Identify and learn the patterns and trends in the production process through the analysis of historical data, and continuously optimize the algorithm parameters; (2) Use machine learning techniques to model the data, identify the key factors affecting production costs, and achieve intelligent decision-making support; (3) Automatically adjust the optimization strategy according to different production scenarios and conditions to improve the overall optimization effect; Conduct dynamic optimization according to the real-time changes in the production environment (such as order adjustment, equipment failure, quality abnormality, etc.), adjust the production plan and resource allocation in real time, and continuously optimize its own parameter settings through the feedback mechanism of the MES system. Feed the optimization results back to each module of the MES system to achieve real-time control of the production process and ensure optimal cost control at all times; The feedback mechanism is used to feed back the optimization results to the MES system in real time, and specifically includes: (1) Automatically generate an optimized production plan and resource allocation plan, and update it to the scheduling module of the MES system; (2) Dynamically adjust the optimization strategy according to the changes in real-time data and market demand, including recalculating the production plan, inventory strategy, and equipment scheduling to ensure the optimal allocation of resources; (3) Provide visual reports and analysis tools to help managers intuitively understand the optimization results and adjustment suggestions and support the decision-making process.

[0018] The following embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation manners and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments. The methods used in the following embodiments are all conventional methods unless otherwise specified.

Claims

1. A cost optimization algorithm module for a manufacturing execution system (MES), characterized in that: It includes a data acquisition unit, a data preprocessing unit, a cost optimization model building unit and an optimization algorithm unit; The data acquisition unit is used to acquire multi-dimensional data related to production, equipment, inventory and quality in real time; The data preprocessing unit is used to clean, normalize and extract features from the multidimensional data acquired by the data acquisition unit; The cost optimization model building unit builds a cost optimization model based on a multi-objective optimization method according to the data preprocessed by the data and processing unit; The optimization algorithm unit combines genetic algorithm and simulated annealing algorithm, adopts intelligent search strategy to solve the model, and generates the optimal production plan and resource allocation plan; The cost optimization algorithm module dynamically updates the optimization results to the manufacturing execution system MES, performs dynamic adjustments and feedback, ensures that the production plan is synchronized with the actual situation, and continuously optimizes the algorithm parameters through historical data analysis and machine learning.

2. The cost optimization algorithm module for a manufacturing execution system MES according to claim 1, characterized in that: The multidimensional data includes raw material procurement costs, real-time inventory information, process parameter data during the production process, equipment operation status data and performance indicators, inventory management data, product quality inspection data, quality control records, customer feedback information, current production schedule and order information; the equipment operation status data includes the equipment's start-up time, downtime, fault records and maintenance history; the product quality inspection data includes the inspection pass rate, rework rate and customer feedback information of each production batch.

3. The cost optimization algorithm module for a manufacturing execution system MES according to claim 1, characterized in that: The cost optimization model construction unit adopts a multi-objective optimization method. The objective functions of the model include minimizing total production costs, minimizing inventory holding costs, maximizing equipment utilization, and maximizing product quality qualification rate. The constraints of the model include production capacity constraints, inventory capacity constraints, equipment availability constraints, and quality standard constraints.

4. The cost optimization algorithm module for a manufacturing execution system MES according to claim 1, characterized in that: The analysis of historical data includes identifying and learning patterns and trends in the production process.

5. The cost optimization algorithm module for a manufacturing execution system MES according to claim 1, characterized in that: The dynamic adjustment and feedback is to perform dynamic optimization according to the real-time changes in the production environment, adjust the production plan and resource allocation in real time, and feed back the optimization results to each module of the manufacturing execution system MES.