Intelligent manufacturing process optimization method and system based on adaptive feedback

Abstract

The invention discloses an intelligent manufacturing process optimization method and system based on adaptive feedback, and relates to the technical field of intelligent manufacturing, an order management platform splits a complex order into a single product type in real time, and ensures the accuracy and traceability of production data; the APS production scheduling system quantifies order values through a profit coefficient, dynamically adjusts production scheduling in combination with a rolling time domain algorithm, intervals predicted processing time and equipment debugging time, and enhances production scheduling elasticity; the production scheduling model creatively fuses three optimization objectives, balances enterprise revenue, equipment load balance and production line total efficiency through normalized weight, and avoids limitation of a single objective. Solving a comprehensive optimal solution under multiple constraint conditions through a genetic algorithm and a simulated annealing algorithm; transparent monitoring is achieved through Gantt chart visualization and a real-time refreshing mechanism, and rapid decision making is assisted; according to the method, dynamic production scheduling, multi-objective optimization and a real-time feedback mechanism are fully fused, and efficient production scheduling in an intelligent manufacturing environment is realized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent manufacturing process optimization method and system based on adaptive feedback. Background Art

[0002] With the rapid development of manufacturing technology, more and more manufacturing companies are adopting intelligent production technology to improve the level of automation and intelligence in production, thereby improving the production efficiency of the company. However, this rapid technological development has also brought new challenges, especially in the management of manufacturing processes.

[0003] With the rapid development of production and manufacturing technology, production scheduling in an intelligent manufacturing environment is extremely important. However, the data integration capability of existing production scheduling is limited, and data silos between different equipment and systems are common, resulting in information barriers and low collaboration efficiency. In addition, the real-time dynamic optimization capability is insufficient, traditional methods rely on preset rules, and the degree of intelligence is uneven. Some companies rely too much on manual experience or a single algorithm model, resulting in insufficient decision-making accuracy and stability, making it difficult to quickly respond to sudden changes in the production environment, and easily causing waste of resources. Production scheduling in the traditional intelligent manufacturing environment restricts the further improvement of the efficiency, flexibility and sustainability of the intelligent manufacturing system.

[0004] In order to solve this problem, we propose an intelligent manufacturing process optimization method and system based on adaptive feedback. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an intelligent manufacturing process optimization method and system based on adaptive feedback, which can effectively solve the problems in the above-mentioned background technology.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is: An intelligent manufacturing process optimization method based on adaptive feedback, comprising: Build an order management platform that receives original orders in real time, splits them into pending orders of a single product type, and assigns unique numbers; Based on the order management platform, obtain the order-related data and corresponding production processes of the orders to be assigned; Build a production management platform, extract all production processes based on the production management platform, and generate a set of related equipment that can be used for processing equipment for each production process; The APS production scheduling system obtains the to-be-allocated orders in real time, splits the to-be-allocated orders into at least one pending order according to the relevant equipment set of the production process corresponding to the to-be-allocated orders, and schedules the orders according to the production process, and calculates the profit coefficient of the pending orders according to the scheduling situation of the APS production scheduling system; Construct a production scheduling model. Based on the production scheduling status of the APS production scheduling system and the profit coefficient of pending orders, the production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process, and maximizing the overall efficiency of the production line as optimization goals, and generates multiple single-objective functions for production scheduling optimization; Generate a comprehensive objective optimization formula based on multiple scheduling optimization single objective functions; According to the production scheduling model, constraints are set and based on the rolling time domain algorithm, the comprehensive target optimization formula is solved to obtain the production scheduling results that meet the constraints, and the production scheduling results are visualized based on the Gantt chart.

[0007] Preferably, the step S1 specifically includes: Acquire equipment-related data and historical production data of the enterprise operation, wherein the historical production data includes a plurality of historical order data and the historical production line data corresponding thereto; The order management platform monitors the order receipt status of the enterprise in real time, obtains the original order information, and generates the to-be-allocated order based on the received original order information. The order management platform splits the original order into at least one to-be-allocated order according to the product type contained in the original order information, and ensures that each generated to-be-allocated order only covers a single product type, assigns a unique order number to the generated to-be-allocated order, and records it as the to-be-allocated order; Obtaining order-related data of the order to be assigned, wherein the order-related data includes product type, production quantity, order time, contract amount, customer level, required delivery time and production material data, wherein the production material data includes processing materials, product specifications and quality standards; Get the corresponding production process according to the product type of the order to be assigned.

[0008] Preferably, the step S3 specifically includes: Extract all production process information in manufacturing from the enterprise's production management platform, the production process information including process name, process type, process description and equipment requirements; According to the equipment-related data, all the extracted production processes are matched with each processing equipment respectively. For each production process, a related equipment set of the processing equipment that can be used is generated. The related equipment set includes all the equipment that can meet the processing requirements of the process, wherein the equipment-related data includes the equipment type, performance parameters, production efficiency, historical usage time, historical equipment failure data and historical equipment maintenance data of each processing equipment; The related device set is represented as: in, Indicates the production process A collection of related devices, Indicates the use in production process No. Processing equipment, Indicates the use in production process The total number of processing equipment.

[0009] Preferably, the step S4 specifically includes: The APS production scheduling system obtains the orders to be allocated in real time, sets the order preparation time point for the orders to be allocated, splits the orders to be allocated into sub-orders with a minimum quantity of 1, and the sum of the production quantities of each sub-order is equal to the production quantity of the order to be allocated; The APS production scheduling system schedules the sub-orders corresponding to the allocated orders and records the sub-orders after scheduling as pending orders; The preparatory production time point is set based on the APS production scheduling system. When the earliest scheduled start time corresponding to the pending order exceeds the preparatory production time point, the schedule of the pending order on the APS production scheduling system is locked and the production scheduling optimization is not performed on the scheduling system, and the pending order is recorded as a processing order; Based on the APS production scheduling system, when the end time of the latest schedule corresponding to the pending order exceeds the current time, the pending order will be recorded as a completed order; Obtain the predicted total production volume of the orders to be allocated in each production process; Obtain equipment-related data of all production equipment in the relevant equipment set corresponding to each production process, and calculate the predicted production efficiency of each production equipment; Based on the order-related data, the predicted total production volume of the pending orders is calculated, and the predicted processing time of the production equipment selected in the generation process of the pending orders is calculated according to the ratio of the production quantity of the pending orders to the production quantity of the pending orders corresponding to the pending orders, as well as the predicted production efficiency of each production equipment; Based on the production scheduling rules of the APS production scheduling system, the predicted equipment debugging time of the production equipment is mapped to the predicted equipment debugging time interval, and the predicted processing time is mapped to the predicted processing time interval. The predicted equipment debugging time interval and the predicted processing time interval are arranged on the Gantt chart of the APS production scheduling system in an adjacent connection relationship; The APS production scheduling system uses a Gantt chart to display the schedule of all processing equipment in the relevant equipment set corresponding to each production process on a horizontal time axis; Based on the production scheduling results of the APS production scheduling system, the profitability coefficient of pending orders is calculated in real time, and the profitability coefficient of orders being processed is determined; Constructing a production scheduling model based on the APS production scheduling system, wherein the production scheduling model optimizes the APS production scheduling system through a defined comprehensive target optimization formula; The calculation formula for the predicted production efficiency is: In the formula, For production process The corresponding related device collection The predicted production efficiency of each production equipment, To improve the production efficiency of production equipment, is the attenuation coefficient, The maintenance times of production equipment within the preset unit time. Preset the number of production equipment failures per unit time, For production process The corresponding related device set The historical usage time of each production equipment; The calculation formula for the predicted processing time is: In the formula, The order to be assigned corresponds to Pending orders in production process The corresponding related device set The predicted processing time of each production equipment is the time in which the order to be assigned is processed in the production process. The total production volume forecasted on The order to be assigned corresponds to Pending orders in production process The predicted production volume on Indicates the order to be allocated Pending orders in production process The predicted production efficiency of the selected production equipment in the corresponding related equipment set, Indicates Production quantity for pending orders, Indicates The production quantity of the pending orders corresponding to the pending orders, The total number of pending orders corresponding to the orders to be allocated; The calculation formula of the profit coefficient of the pending order is: In the formula, Indicates The predicted profit coefficient of pending orders, Indicates The contract amount of the pending orders corresponding to the pending orders, Indicates The material cost of the pending orders corresponding to the pending orders, Indicates of pending orders The energy consumption cost of the production equipment selected for the process.

[0010] Preferably, the construction of the production scheduling model specifically includes: Based on the APS production scheduling system, if the processing equipment is If there is a schedule for the processing equipment at the current time point, the number of the current schedule is recorded as 0, the completed schedules are recorded as -1, -2, -3... in reverse chronological order, and the subsequent scheduled tasks are recorded as 1, 2, 3... in chronological order; if there is no schedule for the processing equipment at the current time point, the number of the first future schedule is recorded as 1, the completed schedules are recorded as 0, -1, -2... in reverse chronological order, and the subsequent scheduled tasks are recorded as 2, 3, 4... in chronological order; With current time As a benchmark, extend backward units of time, will cover The time period is recorded as the prediction cycle time interval; The production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process and maximizing the total efficiency of the production line as optimization goals, and constructs multiple single-objective functions for production scheduling optimization; Generate a comprehensive objective optimization formula based on multiple scheduling optimization single objective functions; The production scheduling model sets constraints and solves the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling results that meet the constraints, and visualizes the production scheduling results based on the Gantt chart; The expression of the single objective function of production scheduling optimization is: In the formula, represents the single objective function of maximizing enterprise benefits, represents the load balancing single objective function, represents the single objective function of maximizing the total efficiency of the production line, Indicates The predicted profit coefficient of the orders being processed / pending, It represents the sum of the number of pending orders and the number of orders being processed. Indicates Customer level coefficient of orders being processed / pending, Indicates The penalty coefficient for unit time delay of orders being processed / pending, Indicates Forecasted delivery time of orders in process / pending, Indicates The stipulated delivery time for the orders in process / pending, Indicates the production process The corresponding related device set The first processing equipment The predicted processing time interval of each schedule is the predicted processing time within the predicted cycle time interval. Indicates the production process The corresponding related device set The first processing equipment The predicted equipment debugging time interval of each schedule is the predicted equipment debugging time within the predicted cycle time interval. Indicates the production process The corresponding related device set The total number of scheduled processing equipment, Indicates the production process Device load balancing weight parameters, Indicates the forecast period duration of the forecast period time interval, Indicates the production process The total number of processing equipment in the corresponding related equipment set; If it does not exist When as well as All are recorded as 0; Among them, when If the production process The corresponding related device set The first processing equipment The orders being processed / pending for the first schedule are If the product type and production material data of the orders being processed / pending in the same schedule are the same, then Take 0; otherwise, Take 1; The constraint condition is expressed as: In the formula, Indicates The order preparation time point of the orders being processed / pending, Indicates The scheduled start time of the first production process in the order being processed / pending, Indicates Production operations in orders in process / pending orders The scheduled end time of Indicates Production operations in orders in process / pending orders Production process The line change time between Indicates Orders being processed / Orders to be processed The scheduled start time of the production process. Indicates The scheduled end time of the last production process in the order being processed / pending, Indicates The time left before delivery of the orders in process / pending, Indicates Forecasted delivery time for orders in process / pending.

[0011] Preferably, the comprehensive objective optimization formula is generated based on multiple production scheduling optimization single objective functions, specifically including: The three single objective functions of production scheduling optimization are normalized by using the range method to obtain the normalized single objective function , and , weight the normalized single objective function and integrate it into a comprehensive objective optimization formula; Calculate the minimum value of the comprehensive target optimization formula, visualize the solution result through the Gantt chart of the production scheduling model, and obtain the production scheduling result that meets the constraint conditions; The calculation formula for the normalization process is: In the formula, Represents a single objective function The maximum value set, Represents a single objective function The minimum value set, where and It can be obtained and set based on historical data; The expression of the comprehensive objective optimization formula is: In the formula, , and Represents the weight parameter.

[0012] Preferably, the method of solving the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling result that meets the constraint conditions and visualizing the production scheduling result based on the Gantt chart specifically includes: The production scheduling model is based on the current time As the starting point, the current time Afterwards it serves as a prediction window; Schedule pending orders as adjustable windows and schedule in-process orders as locked windows; Use heuristic rules to generate the initial scheduling plan for the APS scheduling system; Based on the APS production scheduling system, the genetic algorithm and simulated annealing algorithm are used to calculate the optimal solution of the comprehensive target optimization formula; The production scheduling result that meets the constraint conditions is obtained, and the production scheduling result is visualized based on the Gantt chart.

[0013] The second aspect of the present invention further provides an intelligent manufacturing process optimization system based on adaptive feedback, comprising: The order management platform module is used to receive and split original orders, manage order data, and provide structured input for subsequent production scheduling; The production management platform module is used to manage production processes and equipment resources and generate a collection of related equipment available for production processes; APS production scheduling system module is used to execute order production scheduling through the APS production scheduling system, calculate the profit coefficient, and generate the initial production scheduling plan; The production scheduling optimization module is used to establish a multi-objective optimization model and solve the optimal production scheduling plan; Data feedback and adaptive module, used to update equipment-related data and order-related data in real time, supporting dynamic optimization; The visualization and output module is used to display the production scheduling optimization results based on the production scheduling model and support manual intervention and decision-making.

[0014] Preferably, the order management platform module specifically includes: An order receiving unit is used to receive the original order in real time, split it into orders to be allocated of a single type according to product type, and assign a unique number; The order data extraction unit is used to receive order-related data of the order to be allocated and obtain the production process corresponding to the order to be allocated.

[0015] Preferably, the APS production scheduling system module specifically includes: The order splitting and production scheduling unit is used to build the APS production scheduling system, split the orders to be allocated into sub-orders, record them as pending orders, allocate them to the corresponding production equipment according to the production process, and schedule them; The production scheduling lock unit is used to set the pre-production time point and lock the schedule of the order being processed according to the pre-production time point; Profit coefficient calculation unit, used to calculate the profit coefficient of pending orders based on contract amount, material cost, and energy consumption cost; The Gantt chart generation unit displays the scheduling timeline of each production equipment based on the visual scheduling results.

[0016] Preferably, the production scheduling optimization module specifically includes: The objective function generation unit is used to define three single objective functions: maximizing enterprise benefits, load balancing, and maximizing production line efficiency; Comprehensive objective optimization unit, used to normalize the single objective function and weightedly integrate it into a comprehensive optimization formula; The constraint condition setting unit is used to set the constraint conditions for the enterprise's processing and production; The rolling time domain solving unit is used to solve the comprehensive target optimization formula using genetic algorithm and simulated annealing algorithm, and dynamically adjust the schedule based on the calculation results.

[0017] Compared with the prior art, the present invention provides an intelligent manufacturing process optimization method and system based on adaptive feedback, which has the following beneficial effects: This method dynamically adjusts the production scheduling content through a rolling time domain algorithm based on the APS production scheduling system. Based on the production scheduling situation of the APS production scheduling system, the relevant production scheduling data is input into the production scheduling model. The production scheduling model takes maximizing enterprise benefits, load balancing of each equipment in the production process and maximizing the overall efficiency of the production line as optimization goals, generates multiple production scheduling optimization single objective functions, generates a comprehensive objective optimization formula based on multiple production scheduling optimization single objective functions, uses constraints and optimization rules to solve the comprehensive objective optimization formula, and optimizes the production scheduling model to achieve maximization of enterprise benefits, load balancing of each equipment in the production process and maximization of the overall efficiency of the production line. This method realizes the optimization of the entire chain lean manufacturing process from order reception to production execution through adaptive feedback and intelligent optimization algorithms, providing an efficient, flexible and scalable solution for the digital transformation of the manufacturing industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The present invention is a flowchart of an intelligent manufacturing process optimization method based on adaptive feedback.

[0019] Figure 2 This is a schematic flow chart of step S4 provided by the present invention.

[0020] Figure 3 This is a system framework diagram of an intelligent manufacturing process optimization system based on adaptive feedback according to the present invention. DETAILED DESCRIPTION

[0021] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention will be further described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] In order to address the shortcomings of the existing technology, such as Figure 1 - Figure 3As shown, the present invention provides an intelligent manufacturing process optimization method based on adaptive feedback, comprising: Build an order management platform that receives original orders in real time, splits them into pending orders of a single product type, and assigns unique numbers; Based on the order management platform, obtain the order-related data and corresponding production processes of the orders to be assigned; Build a production management platform, extract all production processes based on the production management platform, and generate a set of related equipment that can be used for processing equipment for each production process; The APS production scheduling system obtains the to-be-allocated orders in real time, splits the to-be-allocated orders into at least one pending order according to the relevant equipment set of the production process corresponding to the to-be-allocated orders, and schedules the orders according to the production process, and calculates the profit coefficient of the pending orders according to the scheduling situation of the APS production scheduling system; Construct a production scheduling model. Based on the production scheduling status of the APS production scheduling system and the profit coefficient of pending orders, the production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process, and maximizing the overall efficiency of the production line as optimization goals, and generates multiple single-objective functions for production scheduling optimization; Generate a comprehensive objective optimization formula based on multiple scheduling optimization single objective functions; Set constraints according to the production scheduling model and solve the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling results that meet the constraints, and visualize the production scheduling results based on the Gantt chart; It should be noted that if Figure 1 As shown, the method dynamically adjusts the production scheduling content through a rolling time domain algorithm based on the APS production scheduling system, and inputs the relevant production scheduling data into the production scheduling model based on the production scheduling situation of the APS production scheduling system. The production scheduling model takes maximizing enterprise benefits, balancing the load of each equipment in the production process and maximizing the total efficiency of the production line as optimization goals, generates multiple production scheduling optimization single objective functions, generates a comprehensive objective optimization formula based on multiple production scheduling optimization single objective functions, uses constraints and optimization rules to solve the comprehensive objective optimization formula, and optimizes the production scheduling model to achieve balancing of maximizing enterprise benefits, balancing the load of each equipment in the production process and maximizing the total efficiency of the production line.

[0023] Specifically, the step S1 specifically includes: Acquire equipment-related data and historical production data of the enterprise operation, wherein the historical production data includes a plurality of historical order data and the historical production line data corresponding thereto; The order management platform monitors the order receipt status of the enterprise in real time, obtains the original order information, and generates the to-be-allocated order based on the received original order information. The order management platform splits the original order into at least one to-be-allocated order according to the product type contained in the original order information, and ensures that each generated to-be-allocated order only covers a single product type, assigns a unique order number to the generated to-be-allocated order, and records it as the to-be-allocated order; Obtaining order-related data of the order to be assigned, wherein the order-related data includes product type, production quantity, order time, contract amount, customer level, required delivery time and production material data, wherein the production material data includes processing materials, product specifications and quality standards; Get the corresponding production process according to the product type of the order to be assigned.

[0024] Specifically, the step S3 specifically includes: Extract all production process information in manufacturing from the enterprise's production management platform, the production process information including process name, process type, process description and equipment requirements; According to the equipment-related data, all the extracted production processes are matched with each processing equipment respectively. For each production process, a related equipment set of the processing equipment that can be used is generated. The related equipment set includes all the equipment that can meet the processing requirements of the process, wherein the equipment-related data includes the equipment type, performance parameters, production efficiency, historical usage time, historical equipment failure data and historical equipment maintenance data of each processing equipment; The related device set is represented as: in, Indicates the production process A collection of related devices, Indicates the use in production process No. Processing equipment, Indicates the use in production process The total number of processing equipment.

[0025] Specifically, Figure 2 As shown, the step S4 specifically includes: The APS production scheduling system obtains the orders to be allocated in real time, sets the order preparation time point for the orders to be allocated, splits the orders to be allocated into sub-orders with a minimum quantity of 1, and the sum of the production quantities of each sub-order is equal to the production quantity of the order to be allocated; The APS production scheduling system schedules the sub-orders corresponding to the allocated orders and records the sub-orders after scheduling as pending orders; The preparatory production time point is set based on the APS production scheduling system. When the earliest scheduled start time corresponding to the pending order exceeds the preparatory production time point, the schedule of the pending order on the APS production scheduling system is locked and the production scheduling optimization is not performed on the scheduling system, and the pending order is recorded as a processing order; Based on the APS production scheduling system, when the end time of the latest schedule corresponding to the pending order exceeds the current time, the pending order will be recorded as a completed order; Obtain the predicted total production volume of the orders to be allocated in each production process; Obtain equipment-related data of all production equipment in the relevant equipment set corresponding to each production process, and calculate the predicted production efficiency of each production equipment; Based on the order-related data, the predicted total production volume of the pending orders is calculated, and the predicted processing time of the production equipment selected in the generation process of the pending orders is calculated according to the ratio of the production quantity of the pending orders to the production quantity of the pending orders corresponding to the pending orders, as well as the predicted production efficiency of each production equipment; Based on the production scheduling rules of the APS production scheduling system, the predicted equipment debugging time of the production equipment is mapped to the predicted equipment debugging time interval, and the predicted processing time is mapped to the predicted processing time interval. The predicted equipment debugging time interval and the predicted processing time interval are arranged on the Gantt chart of the APS production scheduling system in an adjacent connection relationship; The APS production scheduling system uses a Gantt chart to display the schedule of all processing equipment in the relevant equipment set corresponding to each production process on a horizontal time axis; Based on the production scheduling results of the APS production scheduling system, the profitability coefficient of pending orders is calculated in real time, and the profitability coefficient of orders being processed is determined; Constructing a production scheduling model based on the APS production scheduling system, wherein the production scheduling model optimizes the APS production scheduling system through a defined comprehensive target optimization formula; The calculation formula for the predicted production efficiency is: In the formula, For production process The corresponding related device collection The predicted production efficiency of each production equipment, To improve the production efficiency of production equipment, is the attenuation coefficient, The maintenance times of production equipment within the preset unit time. Preset the number of production equipment failures per unit time, For production process The corresponding related device set The historical usage time of each production equipment; The calculation formula for the predicted processing time is: In the formula, The order to be assigned corresponds to Pending orders in production process The corresponding related device set The predicted processing time of each production equipment, For orders to be allocated in production process The total production volume forecasted on The order to be assigned corresponds to Pending orders in production process The predicted production volume on Indicates the order to be allocated Pending orders in production process The predicted production efficiency of the selected production equipment in the corresponding related equipment set, Indicates Production quantity for pending orders, Indicates The production quantity of the pending orders corresponding to the pending orders, The total number of pending orders corresponding to the orders to be allocated; The calculation formula of the profit coefficient of the pending order is: In the formula, Indicates The predicted profit coefficient of pending orders, Indicates The contract amount of the pending orders corresponding to the pending orders, Indicates The material cost of the pending orders corresponding to the pending orders, Indicates of pending orders The energy consumption cost of the production equipment selected for the process; It should be noted that the material cost of the to-be-allocated order corresponding to the to-be-processed order is the purchase price of the processing raw materials of the to-be-allocated order; Specifically, the construction of the production scheduling model specifically includes: Based on the APS production scheduling system, if the processing equipment is If there is a schedule for the processing equipment at the current time point, the number of the current schedule is recorded as 0, the completed schedules are recorded as -1, -2, -3... in reverse chronological order, and the subsequent scheduled tasks are recorded as 1, 2, 3... in chronological order; if there is no schedule for the processing equipment at the current time point, the number of the first future schedule is recorded as 1, the completed schedules are recorded as 0, -1, -2... in reverse chronological order, and the subsequent scheduled tasks are recorded as 2, 3, 4... in chronological order; With current time As a benchmark, extend backward units of time, will cover The time period is recorded as the prediction cycle time interval; The production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process and maximizing the total efficiency of the production line as optimization goals, and constructs multiple single-objective functions for production scheduling optimization; Generate a comprehensive objective optimization formula based on multiple scheduling optimization single objective functions; The production scheduling model sets constraints and solves the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling results that meet the constraints, and visualizes the production scheduling results based on the Gantt chart; The expression of the single objective function of production scheduling optimization is: In the formula, represents the single objective function of maximizing enterprise benefits, represents the load balancing single objective function, represents the single objective function of maximizing the total efficiency of the production line, Indicates The predicted profit coefficient of the orders being processed / pending, It represents the sum of the number of pending orders and the number of orders being processed. Indicates Customer level coefficient of orders being processed / pending, Indicates The penalty coefficient for unit time delay of orders being processed / pending, Indicates Forecasted delivery time of orders in process / pending, Indicates The stipulated delivery time for the orders in process / pending, Indicates the production process The corresponding related device set The first processing equipment The predicted processing time interval of each schedule is the predicted processing time within the predicted cycle time interval. Indicates the production process The corresponding related device set The first processing equipment The predicted equipment debugging time interval of each schedule is the predicted equipment debugging time within the predicted cycle time interval. Indicates the production process The corresponding related device set The total number of scheduled processing equipment, Indicates the production process Device load balancing weight parameters, Indicates the forecast period duration of the forecast period time interval, Indicates the production process The total number of processing equipment in the corresponding related equipment set; It should be noted that the customer grade coefficient is obtained based on the customer grade of the to-be-allocated order corresponding to the order being processed / order to be processed, and the discrete customer grade is converted into a quantifiable normalized coefficient to generate the customer grade coefficient; The unit time delay penalty coefficient represents the The additional cost to be borne per unit time of delay for pending orders / pending orders. The penalty coefficient for delay per unit time is set according to the penalty amount for delay stipulated in the contract. In this embodiment, the penalty coefficient for delay per unit time is set according to the penalty amount for delay per unit time. The penalty amount for delay stipulated in the contract for the pending orders / pending orders corresponding to the pending orders is calculated as 0.1%-0.5% of the contract amount: If it does not exist When as well as All are recorded as 0; Among them, when If the production process The corresponding related device set The first processing equipment The orders being processed / pending for the first schedule are If the product type and production material data of the orders being processed / pending in the same schedule are the same, then Take 0; otherwise, Take 1; The constraint condition is expressed as: In the formula, Indicates The order preparation time point of the orders being processed / pending, Indicates The scheduled start time of the first production process in the order being processed / pending, Indicates Production operations in orders in process / pending orders The scheduled end time of Indicates Production operations in orders in process / pending orders Production process The line change time between Indicates Orders being processed / Orders to be processed The scheduled start time of the production process. Indicates The scheduled end time of the last production process in the order being processed / pending, Indicates The time left before delivery of the orders in process / pending, Indicates Forecasted delivery time of orders in process / pending; Specifically, the comprehensive objective optimization formula is generated based on multiple production scheduling optimization single objective functions, specifically including: The three single objective functions of production scheduling optimization are normalized by using the range method to obtain the normalized single objective function , and , weight the normalized single objective function and integrate it into a comprehensive objective optimization formula; Calculate the minimum value of the comprehensive target optimization formula, visualize the solution result through the Gantt chart of the production scheduling model, and obtain the production scheduling result that meets the constraint conditions; The calculation formula for the normalization process is: In the formula, Represents a single objective function The maximum value set, Represents a single objective function The minimum value set, where and It can be obtained and set based on historical data; The expression of the comprehensive objective optimization formula is: In the formula, , and represents a weight parameter. In this embodiment, , and are set to 0.5, 0.3 and 0.2 respectively, and , and It can be adjusted according to the focus of the optimization goal in corporate operations.

[0026] Specifically, the rolling time domain algorithm is used to solve the comprehensive target optimization formula to obtain the production scheduling result that meets the constraint conditions, and the production scheduling result is visualized based on the Gantt chart, which specifically includes: The production scheduling model is based on the current time As the starting point, the current time Afterwards it serves as a prediction window; Schedule pending orders as adjustable windows and schedule in-process orders as locked windows; Use heuristic rules to generate the initial scheduling plan for the APS scheduling system; Based on the APS production scheduling system, the genetic algorithm and simulated annealing algorithm are used to calculate the optimal solution of the comprehensive target optimization formula; The production scheduling result that meets the constraint conditions is obtained, and the production scheduling result is visualized based on the Gantt chart.

[0027] The second aspect of the present invention also provides an intelligent manufacturing process optimization system based on adaptive feedback, such as Figure 3 As shown, including: The order management platform module is used to receive and split original orders, manage order data, and provide structured input for subsequent production scheduling; The production management platform module is used to manage production processes and equipment resources and generate a collection of related equipment available for production processes; APS production scheduling system module is used to execute order production scheduling through the APS production scheduling system, calculate the profit coefficient, and generate the initial production scheduling plan; The production scheduling optimization module is used to establish a multi-objective optimization model and solve the optimal production scheduling plan; Data feedback and adaptive module, used to update equipment-related data and order-related data in real time, supporting dynamic optimization; The visualization and output module is used to display the production scheduling optimization results based on the production scheduling model and support manual intervention and decision-making.

[0028] Specifically, the order management platform module includes: An order receiving unit is used to receive the original order in real time, split it into orders to be allocated of a single type according to product type, and assign a unique number; The order data extraction unit is used to receive order-related data of the order to be allocated and obtain the production process corresponding to the order to be allocated.

[0029] Specifically, the APS production scheduling system module includes: The order splitting and production scheduling unit is used to build the APS production scheduling system, split the orders to be allocated into sub-orders, record them as pending orders, allocate them to the corresponding production equipment according to the production process, and schedule them; The production scheduling lock unit is used to set the pre-production time point and lock the schedule of the order being processed according to the pre-production time point; Profit coefficient calculation unit, used to calculate the profit coefficient of pending orders based on contract amount, material cost, and energy consumption cost; The Gantt chart generation unit displays the scheduling timeline of each production equipment based on the visual scheduling results.

[0030] Specifically, the production scheduling optimization module includes: The objective function generation unit is used to define three single objective functions: maximizing enterprise benefits, load balancing, and maximizing production line efficiency; Comprehensive objective optimization unit, used to normalize the single objective function and weightedly integrate it into a comprehensive optimization formula; The constraint condition setting unit is used to set the constraint conditions for the enterprise's processing and production; The rolling time domain solving unit is used to solve the comprehensive target optimization formula using genetic algorithm and simulated annealing algorithm, and dynamically adjust the schedule based on the calculation results.

[0031] In summary, the advantages of the present invention are as follows: the order management platform splits complex orders into single product types in real time, simplifies the complexity of production scheduling, and combines unique number tracking to ensure the accuracy and traceability of production data; the production management platform dynamically generates process-equipment matching sets based on equipment performance, maintenance records, etc., to optimize equipment selection; the APS production scheduling system quantifies the order value through the profit coefficient, dynamically adjusts the production scheduling in combination with the rolling time domain algorithm, and compartmentalizes the predicted processing time and equipment debugging time to enhance the flexibility of production scheduling; the production scheduling model innovatively integrates the three major optimization goals, and balances the enterprise benefits, equipment load balancing and the overall efficiency of the production line with normalized weights to avoid the limitation of a single goal; the comprehensive optimal solution under multiple constraints is solved by genetic algorithms and simulated annealing algorithms; the Gantt chart visualization and real-time refresh mechanism realize transparent monitoring and assist in quick decision-making; this method fully integrates dynamic production scheduling, multi-objective optimization and real-time feedback mechanism, and realizes efficient production scheduling in an intelligent manufacturing environment through the deep combination of algorithms and business rules.

[0032] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A smart manufacturing process optimization method based on adaptive feedback, characterized in that: include: S1. Build an order management platform, which receives original orders in real time, splits them into orders to be allocated for a single product type, and assigns unique numbers; S2. Based on the order management platform, obtain the order-related data and corresponding production process of the order to be assigned; S3. Build a production management platform, extract all production processes based on the production management platform, and generate a set of related equipment that can be used for processing equipment for each production process; S4. The APS production scheduling system obtains the to-be-allocated order in real time, splits the to-be-allocated order into at least one pending order according to the relevant equipment set of the production process corresponding to the to-be-allocated order, and schedules it according to the production process, and calculates the profit coefficient of the pending order according to the production scheduling situation of the APS production scheduling system; S5. Construct a production scheduling model based on the production scheduling status of the APS production scheduling system and the profit coefficient of pending orders. The production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process, and maximizing the total efficiency of the production line as optimization goals, and generates multiple single-objective functions for production scheduling optimization; S6. Generate a comprehensive target optimization formula based on multiple production scheduling optimization single objective functions; S7. Set constraints according to the production scheduling model and solve the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling results that meet the constraints, and visualize the production scheduling results based on the Gantt chart.

2. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 1, characterized in that: The step S1 specifically includes: Acquire equipment-related data and historical production data of the enterprise operation, wherein the historical production data includes a plurality of historical order data and the historical production line data corresponding thereto; The order management platform monitors the order receipt status of the enterprise in real time, obtains the original order information, and generates the to-be-allocated order based on the received original order information. The order management platform splits the original order into at least one to-be-allocated order according to the product type contained in the original order information, and ensures that each generated to-be-allocated order only covers a single product type, assigns a unique order number to the generated to-be-allocated order, and records it as the to-be-allocated order; Obtaining order-related data of the order to be assigned, wherein the order-related data includes product type, production quantity, order time, contract amount, customer level, required delivery time and production material data, wherein the production material data includes processing materials, product specifications and quality standards; Get the corresponding production process according to the product type of the order to be assigned.

3. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 2, characterized in that: The step S3 specifically includes: Extract all production process information in manufacturing from the enterprise's production management platform, the production process information including process name, process type, process description and equipment requirements; According to the equipment-related data, all the extracted production processes are matched with each processing equipment respectively. For each production process, a related equipment set of the processing equipment that can be used is generated. The related equipment set includes all the equipment that can meet the processing requirements of the process, wherein the equipment-related data includes the equipment type, performance parameters, production efficiency, historical usage time, historical equipment failure data and historical equipment maintenance data of each processing equipment; The related device set is represented as: in, Indicates the production process A collection of related devices, Indicates the use in production process No. Processing equipment, Indicates the use in production process The total number of processing equipment.

4. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 3, characterized in that: The step S4 specifically includes: The APS production scheduling system obtains the orders to be allocated in real time, sets the order preparation time point for the orders to be allocated, splits the orders to be allocated into sub-orders with a minimum quantity of 1, and the sum of the production quantities of each sub-order is equal to the production quantity of the order to be allocated; The APS production scheduling system schedules the sub-orders corresponding to the allocated orders and records the sub-orders after scheduling as pending orders; The preparatory production time point is set based on the APS production scheduling system. When the earliest scheduled start time corresponding to the pending order exceeds the preparatory production time point, the schedule of the pending order on the APS production scheduling system is locked and the production scheduling optimization is not performed on the scheduling system, and the pending order is recorded as a processing order; Based on the APS production scheduling system, when the end time of the latest schedule corresponding to the pending order exceeds the current time, the pending order will be recorded as a completed order; Obtain the predicted total production volume of the orders to be allocated in each production process; Obtain equipment-related data of all production equipment in the relevant equipment set corresponding to each production process, and calculate the predicted production efficiency of each production equipment; Based on the order-related data, the predicted total production volume of the pending orders is calculated, and the predicted processing time of the production equipment selected in the generation process of the pending orders is calculated according to the ratio of the production quantity of the pending orders to the production quantity of the pending orders corresponding to the pending orders, as well as the predicted production efficiency of each production equipment; Based on the production scheduling rules of the APS production scheduling system, the predicted equipment debugging time of the production equipment is mapped to the predicted equipment debugging time interval, and the predicted processing time is mapped to the predicted processing time interval. The predicted equipment debugging time interval and the predicted processing time interval are arranged on the Gantt chart of the APS production scheduling system in an adjacent connection relationship; The APS production scheduling system uses a Gantt chart to display the schedule of all processing equipment in the relevant equipment set corresponding to each production process on a horizontal time axis; Based on the production scheduling results of the APS production scheduling system, the profitability coefficient of pending orders is calculated in real time, and the profitability coefficient of orders being processed is determined; Constructing a production scheduling model based on the APS production scheduling system, wherein the production scheduling model optimizes the APS production scheduling system through a defined comprehensive target optimization formula; The calculation formula for the predicted processing time is: In the formula, The order to be assigned corresponds to Pending orders in production process The corresponding related device set The predicted processing time of each production equipment, For orders to be allocated in production process The total production forecast on The order to be assigned corresponds to Pending orders in production process The predicted production volume on Indicates the order to be allocated Pending orders in production process The predicted production efficiency of the selected production equipment in the corresponding related equipment set, Indicates Production quantity for pending orders, Indicates The production quantity of the pending orders corresponding to the pending orders, The total number of pending orders corresponding to the orders to be allocated; The calculation formula of the profit coefficient of the pending order is: In the formula, Indicates The predicted profit coefficient of pending orders, Indicates The contract amount of the pending orders corresponding to the pending orders, Indicates The material cost of the pending orders corresponding to the pending orders, Indicates of pending orders The energy consumption cost of the production equipment selected for the process.

5. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 4, characterized in that: The construction of the production scheduling model specifically includes: Based on the APS production scheduling system, if the processing equipment is If there is a schedule for the processing equipment at the current time point, the number of the current schedule is recorded as 0, the completed schedules are recorded as -1, -2, -3... in reverse chronological order, and the subsequent scheduled tasks are recorded as 1, 2, 3... in chronological order; if there is no schedule for the processing equipment at the current time point, the number of the first future schedule is recorded as 1, the completed schedules are recorded as 0, -1, -2... in reverse chronological order, and the subsequent scheduled tasks are recorded as 2, 3, 4... in chronological order; With current time As a benchmark, extend backward units of time, will cover The time period is recorded as the prediction cycle time interval; The production scheduling model takes maximizing enterprise benefits, balancing the load of each device in the production process and maximizing the total efficiency of the production line as optimization goals, and constructs multiple single-objective functions for production scheduling optimization; Generate a comprehensive objective optimization formula based on multiple scheduling optimization single objective functions; The production scheduling model sets constraints and solves the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling results that meet the constraints, and visualizes the production scheduling results based on the Gantt chart; The expression of the single objective function of production scheduling optimization is: In the formula, represents the single objective function of maximizing enterprise benefits, represents the load balancing single objective function, represents the single objective function of maximizing the total efficiency of the production line, Indicates The predicted profit coefficient of the orders being processed / pending, It represents the sum of the number of pending orders and the number of orders being processed. Indicates Customer level coefficient of orders being processed / pending, Indicates The penalty coefficient for unit time delay of orders being processed / pending, Indicates Forecasted delivery time of orders in process / pending, Indicates The stipulated delivery time for the orders in process / pending, Indicates the production process The corresponding related device set The first processing equipment The predicted processing time interval of each schedule is the predicted processing time within the predicted cycle time interval. Indicates the production process The corresponding related device set The first processing equipment The predicted equipment debugging time interval of each schedule is the predicted equipment debugging time within the predicted cycle time interval. Indicates the production process The corresponding related device set The total number of scheduled processing equipment, Indicates the production process Device load balancing weight parameters, Indicates the forecast period duration of the forecast period time interval, Indicates the production process The total number of processing equipment in the corresponding related equipment set; If it does not exist When as well as All are recorded as 0; Among them, when If the production process The corresponding related device set The first processing equipment The orders being processed / pending for the first schedule are If the product type and production material data of the orders being processed / pending in the same schedule are the same, then Take 0; otherwise, Take 1; The constraint condition is expressed as: In the formula, Indicates The order preparation time point of the orders being processed / pending, Indicates The scheduled start time of the first production process in the order being processed / pending, Indicates Production operations in orders in process / pending orders The scheduled end time of Indicates Production operations in orders in process / pending orders Production process The line change time between Indicates Orders being processed / Orders to be processed The scheduled start time of the production process. Indicates The scheduled end time of the last production process in the order being processed / pending, Indicates The time left before delivery of the orders in process / pending, Indicates Forecasted delivery time for orders in process / pending.

6. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 5, characterized in that: The method of generating a comprehensive objective optimization formula based on multiple production scheduling optimization single objective functions specifically includes: The three single objective functions of production scheduling optimization are normalized by using the range method to obtain the normalized single objective function , and , weight the normalized single objective function and integrate it into a comprehensive objective optimization formula; In the formula, Represents a single objective function The maximum value set, Represents a single objective function The minimum value set, where and It can be obtained and set based on historical data; The expression of the comprehensive objective optimization formula is: In the formula, , and Represents the weight parameter.

7. The method for optimizing intelligent manufacturing processes based on adaptive feedback according to claim 6, characterized in that: The method of solving the comprehensive target optimization formula based on the rolling time domain algorithm to obtain the production scheduling result that meets the constraint conditions and visualizing the production scheduling result based on the Gantt chart specifically includes: The production scheduling model is based on the current time As the starting point, the current time Afterwards it serves as a prediction window; Schedule pending orders as adjustable windows and schedule in-process orders as locked windows; Use heuristic rules to generate the initial scheduling plan for the APS scheduling system; Based on the APS production scheduling system, the genetic algorithm and simulated annealing algorithm are used to calculate the optimal solution of the comprehensive target optimization formula; The production scheduling result that meets the constraint conditions is obtained, and the production scheduling result is visualized based on the Gantt chart.

8. An intelligent manufacturing process optimization system based on adaptive feedback, used in an intelligent manufacturing process optimization method based on adaptive feedback as claimed in any one of claims 1 to 7, characterized in that: include: The order management platform module is used to receive and split original orders, manage order data, and provide structured input for subsequent production scheduling; The production management platform module is used to manage production processes and equipment resources and generate a collection of related equipment available for production processes; APS production scheduling system module is used to execute order production scheduling through the APS production scheduling system, calculate the profit coefficient, and generate the initial production scheduling plan; The production scheduling optimization module is used to establish a multi-objective optimization model and solve the optimal production scheduling plan; Data feedback and adaptive module, used to update equipment-related data and order-related data in real time, supporting dynamic optimization; Visualization and output module, used to display the production scheduling optimization results based on the production scheduling model, and support manual intervention and decision-making; The order management platform module specifically includes: An order receiving unit is used to receive the original order in real time, split it into orders to be allocated of a single type according to product type, and assign a unique number; The order data extraction unit is used to receive order-related data of the order to be allocated and obtain the production process corresponding to the order to be allocated.

9. The intelligent manufacturing process optimization system based on adaptive feedback according to claim 8, characterized in that: The APS production scheduling system module specifically includes: The order splitting and production scheduling unit is used to build the APS production scheduling system, split the orders to be allocated into sub-orders, record them as pending orders, allocate them to the corresponding production equipment according to the production process, and schedule them; The production scheduling lock unit is used to set the pre-production time point and lock the schedule of the order being processed according to the pre-production time point; Profit coefficient calculation unit, used to calculate the profit coefficient of pending orders based on contract amount, material cost, and energy consumption cost; The Gantt chart generation unit displays the scheduling timeline of each production equipment based on the visual scheduling results.

10. The intelligent manufacturing process optimization system based on adaptive feedback according to claim 8, characterized in that: The production scheduling optimization module specifically includes: The objective function generation unit is used to define three single objective functions: maximizing enterprise benefits, load balancing, and maximizing production line efficiency; Comprehensive objective optimization unit, used to normalize the single objective function and weightedly integrate it into a comprehensive optimization formula; The constraint condition setting unit is used to set the constraint conditions for the enterprise's processing and production; The rolling time domain solving unit is used to solve the comprehensive target optimization formula using genetic algorithm and simulated annealing algorithm, and dynamically adjust the schedule based on the calculation results.

Images