A production scheduling optimization method and system based on genetic algorithm

Abstract

The application provides a production scheduling optimization method and system based on a genetic algorithm, which comprises the following steps: obtaining required work order data, mold process data and resource data in production scheduling; determining the theoretical total working hours of each work order based on the work order data and the mold process data, and determining the actual available duration of each production device in each shift in the future based on the resource data combined with the production resource calendar; constructing a coding rule with a process priority sequence as the core based on the work order data, the mold process data and the resource data, and initializing a population based on the coding rule and a pre-constructed decoder to obtain an initial scheduling scheme; and performing fitness evaluation and evolution iteration based on the initial scheduling scheme, the theoretical total working hours and the actual available duration to obtain an optimal global scheduling scheme with high device utilization and meeting multiple constraints. The application improves the scientificity and execution efficiency of the scheduling scheme.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a production scheduling optimization method and system based on genetic algorithms. Background Technology

[0002] In manufacturing industries that rely on molds, such as injection molding, stamping, and die casting, production scheduling is a core element that determines production efficiency, directly affecting equipment utilization, production cost control, and order delivery cycle.

[0003] Existing scheduling methods for mold production fall into two categories: manual scheduling and software scheduling based on simple rules. Manual scheduling relies on the planner's experience to allocate and prioritize production tasks. Limited by human computational capabilities, it cannot comprehensively consider multiple constraints such as equipment load, work order priority, and mold changeover costs, easily leading to excessive equipment idle time, frequent mold changeovers, and ultimately low equipment utilization. Software scheduling based on simple rules executes scheduling through preset fixed sorting rules (such as first-come, first-served, shortest processing time priority, etc.). Lacking flexible global optimization logic, it often only satisfies a single optimization objective, sacrificing other key requirements such as work order delivery deadlines and balanced equipment load. Furthermore, its insufficient ability to handle complex production constraints easily leads to production conflicts, ultimately resulting in wasted capacity, increased production costs, and delayed order delivery. In summary, existing scheduling optimization methods generally suffer from technical pain points such as weak constraint handling capabilities, high mold changeover costs, difficulty in balancing multiple optimization objectives, and slow dynamic response. Summary of the Invention

[0004] This invention provides a production scheduling optimization method and system based on genetic algorithms to address the technical pain points of existing scheduling methods, such as weak constraint handling capabilities, high mold change costs, difficulty in balancing multiple objectives, and slow dynamic response, thereby improving the scientific nature and execution efficiency of scheduling schemes.

[0005] In a first aspect, the present invention provides a production scheduling optimization method based on a genetic algorithm, comprising: Based on the ERP system and MES system, obtain the work order data, mold process data and resource data required in the production process; Based on the work order data and the mold process data, the theoretical total working hours of each process corresponding to each work order are determined, and based on the resource data and the production resource calendar, the actual available time of each production equipment in each future shift is determined. Based on the work order data, the mold process data, and the resource data, an encoding rule with the process priority sequence as the core is constructed, and a population initialization is performed based on the encoding rule and the pre-built decoder to obtain an initial scheduling scheme. Based on the initial scheduling scheme, the theoretical total working time of the process, and the actual available time, fitness evaluation and evolutionary iteration are performed to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. Based on the optimal global scheduling scheme, a graphical transformation is performed to obtain a scheduling Gantt chart, and the planner executes production operations according to the scheduling Gantt chart.

[0006] In a second aspect, the present invention also provides a production scheduling optimization system based on a genetic algorithm, applied to the production scheduling optimization method based on a genetic algorithm as described in the first aspect; the production scheduling optimization system based on a genetic algorithm includes: The data acquisition module is used to acquire work order data, mold process data, and resource data required during the production process based on the ERP system and MES system; The calculation and processing module is used to determine the theoretical total working hours of each process corresponding to each work order based on the work order data and the mold process data, and to determine the actual available time of each production equipment in each future shift based on the resource data and the production resource calendar. The initial scheme generation module is used to construct a coding rule based on the work order data, the mold process data and the resource data, with the process priority sequence as the core, and to perform population initialization based on the coding rule and the pre-built decoder to obtain an initial scheduling scheme. The scheme iteration optimization module is used to perform fitness evaluation and evolution iteration based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, so as to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The scheduling visualization and execution module is used to perform graphical transformation based on the optimal global scheduling scheme to obtain a scheduling Gantt chart, and to enable planners to execute production operations based on the scheduling Gantt chart.

[0007] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby implementing the production scheduling optimization method based on genetic algorithms as described above.

[0008] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the production scheduling optimization method based on the genetic algorithm described above.

[0009] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the production scheduling optimization method based on genetic algorithm as described above.

[0010] The production scheduling optimization method based on genetic algorithms provided in this invention obtains work order data, mold process data, and resource data to determine the theoretical total working time of each process corresponding to each work order and the actual available time of each production equipment in each future shift. This clarifies the time benchmark for work order execution and the capacity boundary of the equipment. Based on this data, an initial scheduling scheme that meets the basic production constraints is obtained by constructing an encoding rule with process priority sequence as the core and combining it with a pre-built decoder. This results in a higher quality initial population than random solutions, thereby reducing the search area with large mold change costs, improving the convergence speed and the quality of the final solution. Based on the initial scheduling scheme, the theoretical total working time of the processes, and the actual available time of the equipment, fitness evaluation and evolutionary iteration are performed to achieve global optimization of work order sorting and equipment allocation. This effectively balances multiple objectives such as work order delivery, equipment utilization, and mold change costs, while strengthening the ability to handle complex production constraints. Finally, the optimal global scheduling scheme is transformed into an intuitive scheduling Gantt chart, which facilitates planners to execute production operations efficiently. In summary, this invention effectively solves the technical pain points of existing scheduling methods, such as weak constraint processing capability, high mold change cost, difficulty in balancing multiple objectives, and slow dynamic response, thereby improving the scientific nature and execution efficiency of scheduling schemes. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating the production scheduling optimization method based on genetic algorithms provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the production scheduling optimization system based on genetic algorithm provided in an embodiment of the present invention; Figure 3 An embodiment diagram of the electronic device provided in this invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0014] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0015] See Figure 1 , Figure 1 This is a flowchart illustrating the production scheduling optimization method based on genetic algorithms provided by the present invention. In this embodiment, the execution entity of the production scheduling optimization method based on genetic algorithms is a production scheduling optimization system. Therefore, the production scheduling optimization method based on genetic algorithms includes: Step 10: Obtain the work order data, mold process data, and resource data required in the production process based on the ERP system and MES system.

[0016] Optionally, the production scheduling optimization system establishes a real-time data transmission channel with the Enterprise Resource Planning (ERP) system and Manufacturing Execution System (MES) system connected to production through a preset data interface protocol. Based on the established real-time data transmission channel, and in accordance with preset data acquisition rules and data format requirements, the system automatically extracts three types of core data required for production scheduling, including work order data, mold process data, and resource data. After extraction, the system performs structured processing on the data, removes duplicate and invalid data, and ensures the accuracy and completeness of the data, providing reliable data support for subsequent scheduling calculations.

[0017] Furthermore, the work order data includes the filtered and identified work order number (a string code that uniquely identifies the work order), product code (a category code that uniquely identifies the corresponding product), required quantity (the total number of products that the work order needs to produce), production delivery date (the deadline for completing all production tasks required by the work order), and priority (the work order execution priority is preset based on the urgency of the order and the customer level, divided into 1-5 levels, with level 1 being the highest priority); mold process data includes mold set (the set of molds required to complete the production of the corresponding product, with one product code corresponding to a unique mold set), number of cavities (the number of products that a single mold set can produce in one injection / stamping / die-casting process), standard mold closing and mold change time (standard mold closing time refers to the time it takes for the mold to reach a stable production state after being installed on the equipment, and standard mold change time refers to the time it takes to disassemble one mold set and install and debug another mold set), and pass rate (the historical average number of qualified products when the mold set produces the corresponding product). Product proportion), material and color characteristics (the material type of the raw materials required to produce the product and the surface color attributes of the product; continuous production of work orders with the same material and color characteristics can reduce mold change preparation time); resource data includes mold equipment data and production resource data. Among them, mold equipment data includes equipment number, equipment model, and equipment compatible mold group range (a list of mold groups that a single equipment can be compatible with); production resource data includes equipment status (divided into three states: running, stopped, and maintenance), shift plan (preset equipment production shift arrangements, such as morning shift 8:00-16:00, afternoon shift 16:00-24:00, and evening shift 0:00-8:00), exceptional downtime (statistics on unplanned equipment failure downtime, power outages, and other unforeseen downtime, based on the daily average of the past 30 days of historical data as the basis for prediction), and real-time efficiency (the ratio of the current production rate of the equipment to the theoretical production rate, reflecting the actual operating status of the equipment).

[0018] Step 20: Based on work order data and mold process data, determine the theoretical total working hours of each process corresponding to each work order, and based on resource data combined with the production resource calendar, determine the actual available time of each production equipment in each future shift.

[0019] Optionally, after obtaining the above three types of data, the production scheduling optimization system, based on the required quantity in the work order data and the number of cavities, standard mold closing time, and pass rate in the mold process data, combined with the production cycle characteristics of injection molding / stamping / die casting processes, first calculates the theoretical total working time for each process corresponding to each work order. This theoretical total working time refers to the total time required to complete the production of all required products for that work order under ideal production conditions (no equipment failures, no quality rework, and equipment operating at full capacity). It includes both the standard mold closing time and the actual processing time. The actual processing time also needs to consider the rework and supplementary production needs due to the pass rate. Taking injection molding as an example, the single-mold pure processing cycle is uniformly 2 minutes / cycle. Work order 1: Required quantity 5000 pieces, number of cavities / cycle, pass rate 98%, standard mold closing time 15 minutes, mold change time 10 minutes / cycle. First, calculate the theoretical number of mold production cycles required = required quantity ÷ (number of mold cavities × pass rate) = 5000 ÷ (8 × 98%) ≈ 637.76 times, round up to 638 times. Then calculate the actual processing time = theoretical number of mold production cycles required × single mold processing cycle = 638 times × 2 minutes / time = 1276 minutes. Finally, obtain the theoretical total processing time = standard mold closing time + actual processing time = 15 minutes + 1276 minutes = 1291 minutes. The actual number of qualified products produced = Basic production cycle × Number of cavities × Qualification rate = 638 × 8 × 0.98 ≈ 5002 pieces (2 pieces exceeding the demand, no additional production is needed). It should be noted that if the basic production cycle is 637, the actual number of qualified products = 637 × 8 × 0.98 ≈ 4994 pieces (6 pieces short), then the number of cycles to be supplemented = 6 / (8 × 0.98) ≈ 0.77 → rounded up to 1, then the time for changing the mold during supplementation = Number of cycles × Mold change time = 1 × 10 = 10 minutes (only during supplementation, additional mold change is required; continuous production has no mold change cost). In this case, the theoretical total time of the process needs to be supplemented by the time for changing the mold during supplementation.

[0020] At the same time, the production scheduling optimization system also calculates the actual available time of each production equipment in each future shift based on the equipment status, shift plan, exception downtime, and real-time efficiency in the acquired resource data, combined with the production resource calendar (preset production date range and holiday arrangements, with no holidays within the scheduling period). The actual available time of the equipment refers to the effective time that the equipment can be used for actual processing after deducting necessary non-production time in the corresponding shift. Necessary non-production time includes exception downtime, and the impact of real-time equipment efficiency on actual production capacity must also be considered. Therefore, taking a scheduling period from May 5, 2024 to May 10, 2024 (5 days in total), with no holidays shown on the production resource calendar, as an example, for Equipment 1: the shift plan is morning shift (8:00-16:00, duration 480 minutes) + afternoon shift (16:00-24:00, duration 480 minutes), with an exception downtime of 10 minutes / day as the average possible downtime for each shift, and a real-time efficiency of 95%, then the theoretical duration of a single shift = 480 minutes; the duration of a single shift after deducting exception downtime = 480 minutes - 10 minutes = 470 minutes; the actual available duration of a single shift = duration of a single shift after deducting exception downtime × real-time efficiency = 470 minutes × 95% = 446.5 minutes; finally, the actual available duration per day = actual available duration of morning shift + actual available duration of afternoon shift = 446.5 minutes + 446.5 minutes = 893 minutes.

[0021] Step 30: Based on work order data, mold process data, and resource data, construct coding rules with process priority sequence as the core, and perform population initialization based on coding rules and pre-built decoder to obtain initial scheduling scheme.

[0022] Optionally, the production scheduling optimization system constructs a coding rule based on the obtained work order data, mold process data, and resource data, with the process priority sequence as the core. This coding rule can reflect the work order execution order, and each chromosome in the coding rule corresponds to a set of process priority arrangements. Each gene on the chromosome represents a process corresponding to a work order. The specific coding rule construction is as described in steps 301-305.

[0023] Furthermore, after constructing the encoding rules, the production scheduling optimization system calls a pre-built decoder. The decoder contains built-in mapping rules for parsing these encoding rules. Therefore, the decoder, combined with the constructed encoding rules, inputs the encoding rules into itself. Based on the built-in rules and the encoding rules, the decoder generates an initial scheduling scheme (initial population), essentially transforming the gene sequence into a complete and feasible scheduling scheme. Specifically, the decoder's population initialization process is as follows: First, it parses the sequence order to determine the initial allocation priority order of work orders. For example, the work order at the first priority position is the first allocation target, the second priority position is the second allocation target, and so on, forming an allocation queue. Then, based on the built-in equipment adaptation rules, it selects a set of suitable production equipment for each work order in the allocation queue. For each equipment, it anticipates the actual available time for its future shifts, constructing a candidate pool of equipment shifts for that work order. Finally, the decoder's built-in priority matching rules and time matching rules are used to match the actual available time for each work order in the allocation queue. Single-execution allocation means that from the equipment shift candidate pool of the work order, priority is given to selecting the combination of equipment and shifts whose actual available time is closest to the theoretical total working time of the process and whose actual available time is greater than or equal to the theoretical total working time of the process. If the equipment shift candidate pool cannot satisfy this combination, the shortest consecutive shift combination of equipment with a total available time greater than or equal to the theoretical total working time of the process is selected, and the allocation process of the decoder is repeated to obtain multiple different initial scheduling schemes, forming the initial population of the genetic algorithm. The initial scheduling scheme includes work order allocation details, duration matching details, and corresponding constraint satisfaction indicators.

[0024] Step 40: Based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, perform fitness evaluation and evolutionary iteration to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints.

[0025] Optionally, after obtaining the initial scheduling scheme, the theoretical total working hours of the process, and the actual available time, the production scheduling optimization system first constructs a hierarchical objective function for fitness evaluation, and then performs a genetic iteration operation based on the hierarchical objective function, that is, performs crossover and mutation operations. After the final iteration is completed, the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints is obtained, as shown in steps 401-405.

[0026] Step 50: Based on the optimal global scheduling scheme, a graphical transformation is performed to obtain a scheduling Gantt chart, and the planner executes production operations according to the scheduling Gantt chart.

[0027] Optionally, after obtaining the optimal global scheduling plan, the production scheduling optimization system, based on the optimal global scheduling plan and following the Gantt chart generation rules, transforms the information such as the corresponding process, allocated production equipment, execution time interval (accurate to the minute), and process duration for each work order into a visual scheduling Gantt chart. The Gantt chart uses different colors to distinguish the processes of different work orders and clearly marks the equipment number, time axis, process name, and corresponding work order number. Then, the scheduling Gantt chart is displayed to the planner. After the planner confirms the Gantt chart, the detailed data of the optimal global scheduling plan is written into the database and sent to the MES system, which guides the on-site production personnel to perform the corresponding production operations.

[0028] This invention, through acquiring work order data, mold process data, and resource data, determines the theoretical total working time of each process corresponding to each work order and the actual available time of each production equipment in each future shift. It clarifies the time benchmark for work order execution and the capacity boundary of the equipment. Based on this data, it constructs an encoding rule centered on the process priority sequence and combines it with a pre-built decoder to complete population initialization, obtaining an initial scheduling scheme that meets the basic production constraints. This ensures that the quality of the initial population is higher than that of random solutions, thereby reducing the search area for a large amount of mold change costs, improving the convergence speed and the quality of the final solution. Then, based on the initial scheduling scheme, the theoretical total working time of the processes, and the actual available time of the equipment, it achieves global optimization of work order sorting and equipment allocation through fitness evaluation and evolutionary iteration. This effectively balances multiple objectives such as work order delivery, equipment utilization, and mold change costs, while strengthening the ability to handle complex production constraints. Finally, the optimal global scheduling scheme is transformed into an intuitive scheduling Gantt chart, facilitating efficient production operations by planners. In summary, this invention effectively solves the technical pain points of existing scheduling methods, such as weak constraint processing capability, high mold change cost, difficulty in balancing multiple objectives, and slow dynamic response, thereby improving the scientific nature and execution efficiency of scheduling schemes.

[0029] In one embodiment, the process of steps 301-305 includes Step 301: Based on the product code and required quantity in the work order data and the mold group, number of mold cavities, and pass rate in the mold process data, determine the basic parameters for process adaptation between the work order and the mold group, and determine the process adaptation index corresponding to each work order based on the basic parameters for process adaptation; the process adaptation index characterizes the degree of process matching between the work order and the corresponding mold group.

[0030] Optionally, the production scheduling optimization system integrates the product code and required quantity from the acquired work order data, as well as the mold group, number of cavities, and pass rate from the mold process data, into basic parameters for process adaptation between work orders and mold groups. Specifically, these parameters can be divided into: product mold unique matching parameters (i.e., one product code corresponds to one unique mold group, with a value of 1); and cavity adaptation parameters. This refers to the degree of matching between the required quantity of work order j and the corresponding number of mold cavities, normalized to the interval [0.1, 1] by taking the natural logarithm of the ratio of required quantity to number of mold cavities; the pass rate matching parameter. The pass rate (range 0-1) from the mold process data is used to characterize the quality stability of the product produced by the mold group for this work order. Therefore, a comprehensive index characterizing the degree of process matching between work order j and the corresponding mold group is calculated using a nonlinear coupling formula, the specific formula of which is as follows: ;in, is the process matching index for work order j, with a value range of (0, 1]. The larger the value, the higher the degree of process matching. Let j be the number of work orders required.

[0031] Step 302: Based on the process adaptability index and the production delivery date and priority in the work order data, and the equipment status and shift plan in the production resource data, determine the adaptation correlation parameters between the work order and the production resources, and determine the resource availability index corresponding to each work order based on the adaptation correlation parameters; the resource availability index characterizes the degree of immediate availability matching of the production resources corresponding to the work order.

[0032] Optionally, the production scheduling optimization system, based on the obtained process fit index, combined with the production delivery date and priority in the work order data, and the equipment status and shift plan in the production resource data, determines the fit correlation parameters between work and production resources, and then calculates the resource availability index. These fit correlation parameters include delivery urgency parameters. This refers to the ratio of the remaining production time of work order j to the theoretical total working hours of the process. The smaller the ratio, the more urgent the delivery date. Remaining production time = production delivery date - scheduling start time; priority weight parameter. This refers to the weight value set based on the work order priority. For example, if the priority obtained in step 10 is divided into 1-5 levels, the corresponding weight parameters are 1.0, 0.8, 0.6, 0.4, and 0.2 (level 1 is the highest, with a weight of 1.0); available equipment parameters. This refers to the comprehensive availability value of the equipment adapted to the mold group corresponding to work order j. It is calculated by multiplying the percentage of equipment in operation among the adapted equipment by (1 - percentage of exceptional downtime), where the percentage of exceptional downtime = average daily exceptional downtime / total daily shift duration per equipment. Then, based on resource availability indicators... The degree of immediate availability matching of production resources corresponding to work orders is characterized by a nonlinear product formula, which strengthens the coupling effect between the urgency of delivery parameters and the availability of equipment parameters. The formula is as follows: ;in, is the resource availability index for work order j, with a value range of (0, 1]. The larger the value, the higher the degree of real-time resource matching.

[0033] Step 303: Based on the process adaptability index and resource availability index, construct the basic priority coefficient for each work order.

[0034] Optionally, after determining the process suitability index and resource availability index, the production scheduling optimization system constructs a basic priority coefficient for each work order. This coefficient serves as the benchmark for initially defining the processing priority ranking of work orders. By integrating the core influences of process matching degree and resource availability, it is calculated using a non-linear coupling formula, and the priority weight of work orders with high suitability and high availability is strengthened through an exponential function. The formula is as follows: ;in, is the basic priority coefficient for work order j, with a value range of (0, 1]. The larger the value, the higher the initial processing priority of the work order.

[0035] Step 304: Based on the basic priority coefficient and the material and color characteristics and standard mold change time in the mold process data, determine the mold change association parameters of the work order, and determine the mold change adaptation correction coefficient for each work order based on the mold change association parameters; the mold change adaptation correction coefficient is used to correct the mold change process influence not considered in the basic priority coefficient.

[0036] Optionally, after determining the basic priority coefficient, the production scheduling optimization system combines it with the material and color characteristics and standard mold changeover time from the acquired mold process data to determine the work order mold changeover correlation parameters. Then, it calculates the mold changeover adaptation correction coefficient to correct for mold changeover process impacts not considered in the basic priority coefficient, thereby optimizing mold changeover costs. These work order mold changeover correlation parameters include material and color matching parameters. Parameters affecting mold change time Among them, material color matching parameters To determine the degree of matching between work order j and other work orders in terms of material and color characteristics, if at least one other work order has the same material and color characteristics as work order j, then... (Indicating continuous production and reduced mold changes), otherwise (Characteristics require separate mold change); mold change time affects parameters. The normalized value of the standard mold changeover time for the mold group corresponding to work order j is calculated as (maximum standard mold changeover time - standard mold changeover time for this work order) / (maximum standard mold changeover time - minimum standard mold changeover time), normalized to the interval [0.8, 1.2]. The shorter the mold changeover time, the larger the parameter value. Then, a coupled formula with the basic priority coefficient is used to strengthen the correction effect of the mold changeover process on the priority. The formula is: ;in, is the mold change adaptation correction coefficient for work order j, with a value range of (0, 2]. The larger the value, the better the mold change process adaptability and the more positive the correction magnitude to the basic priority.

[0037] Step 305: Based on the basic priority coefficient and the mold change adaptation correction coefficient, calculate the target priority value of each work order, and sort them from largest to smallest according to the target priority value to obtain the coding rule with the process priority sequence as the core.

[0038] Optionally, the production scheduling optimization system calculates the target priority value for each work order based on the determined basic priority coefficient and mold change adaptation correction coefficient. This target priority value comprehensively integrates process adaptability, resource availability, and mold change process adaptability, serving as the core quantitative basis for coding rules. This ensures that the priority ranking conforms to production constraints while optimizing mold change costs and equipment utilization. The formula is as follows: ;in, Let be the target priority value of work order j, with a value range of (0, +∞). The larger the value, the higher the final processing priority of the work order. Furthermore, the formula introduces an exponential function to amplify the priority difference of work orders with high coefficients, so as to avoid scheduling conflicts caused by priority ambiguity.

[0039] Furthermore, after determining the target priority value for each work order, all work orders are sorted from largest to smallest according to their target priority values ​​to form a process priority sequence. This sequence is the core of the coding rule. The order of work orders in the sequence corresponds to the order of chromosome genes, and each work order corresponds to a gene identifier, ultimately forming a complete coding rule with the process priority sequence as its core.

[0040] This invention constructs a multi-dimensional, quantifiable priority evaluation system that integrates core constraints and objectives of the mold industry, such as process adaptability, resource availability, and mold change costs, avoiding the limitations of single-rule sorting. It uses the process priority sequence as the core, precisely matching it with the indirect encoding and parsing logic of the subsequent decoder to ensure the feasibility and efficiency of initial population generation. Furthermore, through hierarchical calculation of indicators and gradual correction of coefficients, the encoding rules reflect both the inherent processing priority of the work order and the mold industry's unique mold change cost control requirements, providing a high-quality encoding foundation for subsequent global optimization by the genetic algorithm.

[0041] In one embodiment, the process of steps 401-405 includes: Step 401: Based on the theoretical total working time of the process, the actual available time, and the initial scheduling plan, constraint mapping is performed to obtain the rigid constraints and flexible constraints of each process. Based on the rigid constraints and flexible constraints, a hierarchical objective function combining the primary objective requirement and the secondary objective requirement is constructed. The rigid constraints include time window constraints and equipment matching constraints; the flexible constraints are priority constraints; the primary objective requirement is equipment utilization rate; and the secondary objective requirements include total mold change time, on-time order delivery rate, and work order priority.

[0042] Optionally, the production scheduling optimization system performs constraint mapping processing based on the determined theoretical total working time of the process, the actual available time of the production equipment, and the initial scheduling scheme generated in step 30. The processing process is as follows: First, extract the hard constraints of each process in terms of time and equipment, and determine them as rigid constraints, including time window constraints (i.e., the execution time interval of the work order must fall completely within the actual available time of the corresponding shift of the equipment, and the actual start time of the work order ≥ the scheduling start time, the actual end time ≤ the production delivery date, and the execution time of the work order ≥ the theoretical total working time of the process) and equipment matching constraints (i.e., the mold group corresponding to the process can only be assigned to the production equipment in the compatible equipment list); then extract the elastic constraints related to the work order priority, and determine them as flexible constraints (here, priority constraints, i.e., high-priority work orders should be allocated resources first, and low-priority work orders can adjust their execution order without affecting the main objective). Subsequently, based on the determined rigid and flexible constraints, a hierarchical objective function combining primary and secondary objective requirements is constructed, as detailed in steps 4011-4014. The primary objective requirement is equipment utilization rate, i.e., maximizing equipment utilization rate. The secondary objective requirements include total mold changeover time, on-time order delivery rate, and work order priority.

[0043] Step 402: Calculate the primary and secondary objective values ​​of each scheme in the initial scheduling scheme based on the hierarchical objective function to obtain the scheme objective value set.

[0044] Optionally, the production scheduling optimization system calculates the primary objective value (equipment utilization rate) and secondary objective values ​​(comprehensive total mold changeover time ratio, on-time order delivery rate, and work order priority satisfaction rate) for each determined initial scheduling scheme (i.e., initial population) based on the determined hierarchical objective function. The primary objective value and all secondary objective values ​​of the same scheme are integrated into a complete set of scheme objective values ​​to ensure that the optimization effect of each scheme is quantifiable and comparable. During the calculation process, it verifies whether the scheme meets the rigid constraints. Only the effective schemes that meet the rigid constraints are calculated for objective values, and the invalid schemes (those that violate the rigid constraints) are directly eliminated and not included in the subsequent evaluation.

[0045] Step 403: Based on the set of scheme target values, sort the schemes in descending order according to the primary target value, divide the schemes whose primary target value difference is less than a preset value into the same primary target group, sort the schemes in the same primary target group according to the secondary target value to obtain the secondary target non-dominated level, and calculate the secondary target crowding distance of the schemes in the same primary target group to obtain the fitness evaluation result of each scheme.

[0046] Optionally, the production scheduling optimization system, based on the obtained set of target values, sorts the primary target values ​​in descending order to prioritize the optimization of the primary target (equipment utilization). A preset threshold of 1% is set, and schemes with a primary target value difference of less than 1% are grouped into the same primary target group. Then, a non-dominated ranking is performed on the schemes within the same primary target group, meaning that no other scheme in the group has a better secondary target value, and at least one secondary target dimension is superior. After ranking, a non-dominated hierarchy is obtained: the first level represents the optimal non-dominated scheme within the group, the second level represents schemes dominated by the first level, and so on. Finally, for schemes within the same primary target group, the secondary target congestion distance is calculated to reflect the dispersion of the scheme in the secondary target space. The formula is as follows: Where j is the index of the sorted schemes within the group; , This represents the secondary objective value of adjacent schemes; , The maximum and minimum values ​​of the secondary objectives within the group are set; the crowding distance of the boundary schemes (first and last indices) is set to 1 (maximum value); finally, the fitness evaluation results of each scheme are obtained, including the overall priority of the primary objective group (the higher the primary objective value within the group, the higher the priority), the non-dominated level of the secondary objectives (the lower the level, the higher the priority), and the crowding distance (the larger the distance, the higher the priority), to obtain the fitness ranking of each scheme. The higher the ranking, the better the fitness.

[0047] Step 404: Based on the fitness evaluation results, the initial scheduling schemes are screened and retained to obtain an elite scheme set. The remaining schemes in the initial scheduling schemes are then screened using the roulette wheel selection method according to the order of primary objective grouping priority and secondary objective crowding distance to obtain a set of schemes to be crossovered.

[0048] Optionally, the production scheduling optimization system performs a dual screening of the initial scheduling schemes based on the obtained fitness evaluation results, resulting in an elite scheme set and a crossover scheme set, thus providing high-quality parent individuals for subsequent evolutionary iterations. Specifically, in the elite scheme set screening, the top 20% of schemes in terms of fitness are selected to form the elite scheme set, which is directly retained to the next generation population to ensure that high-quality genes are not lost. That is, the number of elite schemes = the number of effective schemes × 20%, rounded up. For the crossover scheme set screening, the remaining 80% of schemes are selected according to the order of primary target group priority → secondary target crowding distance, and a roulette wheel selection method is used to select schemes equal to the number of elite schemes to form the crossover scheme set. The selection probability allocation rules are: the higher the priority of the primary target group, the greater the selection probability weight of the schemes within the group; within the same primary target group, the greater the secondary target crowding distance, the higher the selection probability.

[0049] Step 405: Based on the elite scheme set and the scheme set to be crossed, perform crossover, mutation and iterative termination of the population to obtain the optimal global scheduling scheme.

[0050] Optionally, after determining the elite solution set and the solution set to be crossed, the production scheduling optimization system merges the elite solution set and the solution set to be crossed into the current population, and performs population evolution operations (including crossover and mutation) and iterative update operations until the iteration terminates, to obtain the optimal global scheduling solution, as shown in steps 4051-4054.

[0051] This invention precisely defines the core constraints of mold production through constraint mapping, ensuring that the constructed hierarchical objective function not only guarantees the primary goal of maximizing equipment utilization but also takes into account the comprehensive optimization of secondary objectives. By combining fitness evaluation with primary objective grouping, non-dominated sorting, and crowding distance, it ensures the differentiation between good and bad solutions while maintaining population diversity and preventing the algorithm from getting stuck in local optima. Furthermore, through a screening strategy combining elite retention and roulette wheel selection, as well as targeted crossover and mutation operators, it ensures the inheritance of superior genes and global search capabilities. The final optimal global scheduling solution generated is both feasible and optimizable.

[0052] In one embodiment, the process of steps 4011-4014 includes: Step 4011: Based on the feasibility constraints in the mold scheduling scenario, construct a feasibility verification layer with the constraint compliance judgment rules of the scheduling scheme as the core, and screen the initial scheduling scheme within the feasibility verification layer to obtain a set of feasible scheduling schemes.

[0053] Optionally, the production scheduling optimization system constructs a feasibility verification layer based on the feasibility constraints in the mold scheduling scenario (i.e., the rigid constraints determined in step 401), with the constraint compliance judgment rules of the scheduling scheme as its core. These feasibility constraints are the time window constraints and equipment matching constraints of the rigid constraints determined in step 401. The logic of these constraint compliance judgment rules is as follows: if a scheduling scheme fully satisfies all rigid constraints, it is judged as compliant and feasible; if any rigid constraint is violated, it is judged as non-compliant and infeasible. Based on this, the initial scheduling schemes are verified one by one, and all compliant and feasible schemes are selected to form a set of feasible scheduling schemes.

[0054] Step 4012: Based on the set of feasible scheduling schemes, combined with the actual available time of equipment and the theoretical total working time of the process, determine the actual production time of equipment corresponding to each feasible scheduling scheme. Based on the actual production time of equipment, construct a core target layer with equipment utilization rate as the core. In the core target layer, sort and filter the feasible scheduling schemes to obtain a subset of feasible scheduling schemes with similar equipment utilization rates after sorting.

[0055] Optionally, the production scheduling optimization system extracts the process information allocated to each production equipment in each feasible scheduling scheme based on the set of feasible scheduling schemes obtained in step 4011. Combining this with the actual available time of the equipment and the theoretical total working time of the process calculated in step 20, it determines the actual production time of the equipment corresponding to each feasible scheduling scheme (i.e., the sum of the total time for all production equipment in the scheme to execute the allocated processes). Then, a core objective layer is constructed with maximizing equipment utilization as the core. The core objective layer includes equipment utilization calculation rules and scheme sorting and filtering rules: equipment utilization is calculated according to a preset formula (i.e., equipment utilization = actual production time of equipment / total actual available time of equipment), and the sorting and filtering rules are sorted in descending order of equipment utilization. The equipment utilization difference threshold is set to 1%, and schemes with a difference of less than 1% in equipment utilization after sorting are divided into a subset of feasible scheduling schemes with similar equipment utilization.

[0056] Step 4013: Based on the subset of feasible scheduling schemes, combined with work order priority, on-time delivery rate of orders and mold changeover time, determine the total mold changeover time and high-priority work order delivery status for each scheme, and construct the secondary target layer based on the total mold changeover time and high-priority work order delivery status.

[0057] Optionally, the production scheduling optimization system, based on the subset of feasible scheduling schemes obtained in step 4012, and combined with the obtained work order priorities, production delivery dates, and mold changeover times, first determines the total mold changeover time and the delivery status of high-priority work orders for each scheme. Then, it constructs a secondary objective layer containing three sub-objectives. In this secondary objective layer, secondary objective values ​​are calculated to comprehensively reflect the optimization requirements of minimizing the total mold changeover time, maximizing the on-time delivery rate of high-priority work orders, and maximizing the work order priority matching degree. Here, the total mold changeover time is the sum of the mold changeover times generated during the execution of processes by all equipment in the scheme. The delivery status of high-priority work orders refers to the indicator of whether high-priority work orders are delivered within the production delivery date; 1 indicates completion, and 0 indicates non-completion. The final secondary objective function of the constructed secondary objective layer is: ; ;in, On-time delivery identifier for high-priority work orders (On-time delivery identifier) ,otherwise ); This represents the total number of work orders in the subset of feasible scheduling schemes; The priority matching degree of the work order; Let the priority level weight of work order j be (e.g., level 1 = 5, level 2 = 3, level 3 = 1, then) ); For the plan The sum of mold-changing times for all equipment in the process; minutes (preset maximum mold change time); The range of values ​​for the secondary objective is (0, +∞), and the larger the value, the better the optimization effect of the secondary objective. For scheduling scheme The actual weight of the gene locus in the middle work unit j, i.e., the scheduling scheme. The value in the middle represents the gene locus assigned to the work order, ranging from [1, 10], and must meet the corresponding weight lower limit (Level 1 ≥ 8, Level 2 ≥ 5, Level 3 ≥ 2), which can be determined according to the scheduling scheme. Chromosome encoding acquisition.

[0058] Step 4014: Integrate the feasibility verification layer, core objective layer, and secondary objective layer to obtain the hierarchical objective function.

[0059] Optionally, the production scheduling optimization system integrates the feasibility verification layer constructed in step 4011, the core objective layer constructed in step 4012, and the secondary objective layer constructed in step 4013 according to a hierarchical and progressively decreasing priority logic to obtain a hierarchical objective function. The hierarchical execution logic of this function is as follows: the first layer executes the feasibility verification layer rules to eliminate non-compliant solutions; the second layer executes the core objective layer rules on compliant solutions, sorting them by equipment utilization and filtering a subset of similar solutions; the third layer executes the secondary objective layer rules on the subset of similar solutions, sorting them according to the obtained secondary objective values. The layers strictly adhere to the constraint that the next layer will not proceed until the previous layer is completed, ensuring the optimization logic of prioritizing primary objectives and supplementing with secondary objectives.

[0060] This invention, through its hierarchical objective function, strictly adheres to the logic of feasibility as a prerequisite, maximizing equipment utilization as the core, and secondary objective optimization as a supplement. This precisely aligns with the core demands of production scheduling in the mold industry, avoiding the problem of reversed priorities in multi-objective optimization. Furthermore, by eliminating illegal solutions in advance through a feasibility verification layer, the computational load of subsequent optimization is reduced, improving algorithm efficiency. The progressive design of the core objective layer and the secondary objective layer not only ensures the priority of the primary objective but also effectively distinguishes similar solutions through refined secondary objective indicators, solving the pain points of existing scheduling methods that are difficult to balance multiple objectives and have inaccurate constraint handling.

[0061] In one embodiment, the process of steps 4051-4054 ​​includes: Step 4051: Based on the coding rules of the set of cross-schemes and the initial scheduling scheme, determine the process priority coding string of the scheme.

[0062] Optionally, the production scheduling optimization system determines the process priority encoding string for each crossover scheme based on the obtained set of schemes to be crossovered and the encoding rules for generating the initial scheduling scheme in step 30 (such as the gene position correspondence and weight lower limit constraints of work order 2-equipment 2, work order 1-equipment 2, work order 1-equipment 1, and work order 3-equipment 1). This process priority encoding string directly reflects the priority weight allocation of the work order-equipment combination and is a concrete expression of the scheduling scheme chromosome. Furthermore, the process priority encoding string is a 4-bit integer sequence. , where each integer The specific correspondence between a work order and a device combination in the corresponding coding rule is as follows: Work order 2 - Equipment 2 (Level 1 priority, lower weight limit 8); Work order 1 - Equipment 2 (Level 2 priority, lower weight limit 5); Work order 1 - Equipment 1 (Level 2 priority, lower weight limit 5); Work order 3 - Device 1 (priority level 3, weight lower limit 2). The encoding string constraint requires that each bit in the encoding string be an integer. The following conditions must be met: ① Value range [1, 10]; ② No repetition of the 4-digit integers; ③ Compliance with the lower limit constraint of the weight of the corresponding work order-equipment combination. The code string extraction process is as follows: For each scheme to be cross-referenced, extract the 4-digit weight value from its chromosome code, and form a process priority code string according to the above correspondence, ensuring that the code string is completely consistent with the priority allocation logic of the original scheme.

[0063] Step 4052: Based on the process priority encoding string, swap the process priority segments between any two intersection points in the set of crossover schemes, and retain the process priority order and equipment-process matching correlation outside the intersection point to obtain the preliminary crossover scheme.

[0064] Optionally, the production scheduling optimization system, based on the obtained process priority coding strings, pairs the crossover schemes within the same main target group, exchanging coding string segments using a two-point crossover method. Simultaneously, it preserves the process priority order and equipment-process matching correlation outside the crossover point, generating preliminary crossover schemes. Specifically, the core of this crossover operation is to integrate the superior priority allocation genes of two parent schemes while maintaining the work order-equipment matching constraint. The pairing rules are as follows: Schemes to be crossover are randomly paired in pairs according to the main target group priority from high to low, ensuring priority crossover of schemes within the same group (reducing gene fusion of schemes with large differences in the main target, avoiding population degradation); if the number of schemes to be crossover is odd, the last scheme is paired with a scheme randomly selected from the elite scheme set; in the selection of crossover points: for each pair of coding strings, a random number generator selects two non-repeating crossover points (index). , , The segments between the intersection points are the areas to be exchanged; and during the segment exchange: the segments between the two parent encoding strings are exchanged to form two temporary encoding strings; finally, constraint repair and preliminary crossover scheme determination are performed: that is, constraint verification (numerical uniqueness, weight lower limit) is performed on the temporary encoding strings. If there are duplicate values, they are replaced with unused integers in [1, 10] that meet the weight lower limit; the repaired encoding strings correspond to the generation of preliminary crossover schemes. The work order-equipment matching relationship of this scheme is consistent with the encoding rules, only the priority weight is adjusted.

[0065] Step 4053: Based on the preliminary crossover scheme and the utilization rate and idle time information of each device determined by the actual available time, a variation operation is performed in combination with the theoretical total working time of the process to obtain the variation scheme.

[0066] Optionally, the production scheduling optimization system first determines the utilization rate and idle time information of each piece of equipment based on the determined preliminary crossover scheme and the actual available time. Then, based on the determined utilization rate and idle time information of each piece of equipment, it performs a mutation operation to finally obtain the mutation scheme, as shown in steps 40531-40534.

[0067] Step 4054: Based on the elite solution set, variant solutions, theoretical total working time of the process and actual available time, iterative updates and solution selection are performed to obtain the optimal global scheduling solution.

[0068] Optionally, after obtaining the elite solution set, variant solutions, theoretical total working time of the process, and actual available time, the production scheduling optimization system merges the elite solution set from step 404 with the variant solutions from step 4053 to form a new generation population. Then, it iterative updates and solution selection are performed based on the theoretical total working time of the process and the actual available time of the equipment until the iteration termination condition is met, and the optimal global scheduling solution is output, as shown in steps 40541-40544.

[0069] This invention provides a standardized carrier for crossover and mutation operations through process priority encoding strings, ensuring the uniformity and standardization of evolutionary operations. Crossover operations retain the matching correlation between equipment and processes, while mutation operations combine equipment idle time and the theoretical total working time of processes, ensuring the feasibility of the solution during the evolutionary process and avoiding the generation of illegal solutions. Finally, in the iterative update process, the excellent genes of elite solutions and the innovative characteristics of mutation solutions are integrated, taking into account the stability and diversity of the population and improving the global search capability of the algorithm. The optimal global scheduling solution selected in the end achieves comprehensive optimization in terms of equipment utilization, mold change cost control, and order delivery guarantee, effectively solving the pain points of difficulty in balancing multiple objectives and weak dynamic optimization capability in mold production scheduling.

[0070] In one embodiment, the process of steps 40531-40534 includes: Step 40531: Based on the preliminary cross scheme, the theoretical total working time of the process and the actual available time, determine the utilization gap value of each equipment and the cross-equipment mold change saving potential of each process, and construct a pool of variant candidate processes based on the utilization gap value and the cross-equipment saving potential; the cross-equipment mold change saving potential refers to the difference in mold change time between different equipment.

[0071] Optionally, the production scheduling optimization system, based on the determined preliminary crossover scheme, the theoretical total working time of each process, and the actual available time, first calculates the utilization gap value of each piece of equipment and the cross-equipment mold change saving potential of each process. The formula for calculating the utilization gap value is as follows: ;in, The actual production time of device i in the preliminary crossover scheme. This refers to the actual available time of device i; The percentage of unutilized available capacity of equipment i, ranging from [0, 1], indicates that the equipment is more idle and has greater optimization potential. In calculating the potential savings from cross-equipment mold changes, for processes (work orders) that can be produced across equipment, the difference in mold change time between different compatible equipment is calculated using the following formula: ( , (This refers to the set of devices that can be adapted to work order j). To adapt the mold change time of work order j to equipment i, The larger the value, the greater the potential to save mold change time through equipment replacement.

[0072] Next, a candidate process pool for mutations is constructed according to preset screening rules. This pool focuses on processes with room for equipment replacement and optimization, providing precise candidate objects for subsequent mutation operations. Specifically, the screening rules for candidate processes are: single-condition screening: the work order corresponding to the process must meet the utilization gap value. (Corresponding to significant equipment idle time) or potential for cost savings through cross-equipment mold changing. Minutes (significant optimization space for mold replacement); Multi-condition verification: The selected process must simultaneously satisfy the following conditions: the theoretical total working time of the process after equipment replacement ≤ the remaining idle time of the target equipment, and the target equipment and work order j are compatible (meeting equipment matching constraints); Finally, the processes that meet the above rules are included in the variant candidate process pool, which records key information such as the current allocated equipment, the list of compatible equipment, the utilization gap value, and the potential for saving costs when changing molds across equipment for each candidate process.

[0073] Step 40532: Based on the remaining idle time, cumulative mold change count, and current load rate of each matchable device within the time window of each candidate process determined by the mutated candidate process pool and the actual available time, construct the dynamic adaptability of the device to the candidate process.

[0074] Optionally, the production scheduling optimization system first determines the remaining idle time, cumulative mold change count, and current load rate of each matching equipment within the time window of each candidate process based on the determined pool of candidate processes and the actual available time. Then, based on the obtained remaining idle time, cumulative mold change count, and current load rate of each matching equipment, it constructs the dynamic adaptability of the equipment to the candidate processes. This dynamic adaptability comprehensively reflects the matching optimization potential between the equipment and the candidate processes and is the core decision basis for equipment replacement.

[0075] Specifically, in determining the remaining idle time, the remaining idle time... Equipment In candidate process Within the time window (process) The available time (from the planned start time to the planned end time), and ;in, For equipment Total duration of processes allocated within this time window; cumulative number of mold changes. Refers to equipment In candidate process The number of planned mold changes within the time window; the current load rate. Its value ranges from [0, 1], with a larger value indicating a heavier current load on the device. Based on this, a nonlinear dynamic adaptability formula coupled with multiple dimensions is adopted to strengthen the synergistic effect of remaining idle time, number of mode changes, and load rate. The dynamic adaptability formula is as follows: ;in, For equipment For candidate processes The dynamic adaptability, with a value range of (0, 1], indicates that the larger the value, the more suitable device i is to undertake the candidate process. ; This refers to the maximum available time of the equipment per shift (e.g., 8 hours = 480 minutes), or the maximum daily working time of the equipment preset according to the production plan; and if (i.e., the device's idle time exceeds the maximum available time), then This ensures that the value range is always within (0, 1). Then, for each candidate process in the mutation candidate process pool, all its matching devices are traversed, the key parameters of each device are calculated, and the dynamic fit degree is obtained by substituting them into the dynamic fit degree formula, thus forming a process-device-fit degree mapping table.

[0076] Step 40533: Based on the dynamic adaptability and the process equipment allocation relationship of the preliminary cross scheme, determine the current allocated equipment and the optimal equipment for each candidate process, and replace the equipment according to the criteria of filling the utilization gap and saving mold replacement based on the current allocated equipment and the optimal equipment to obtain the transition scheme after equipment allocation variation.

[0077] Optionally, the production scheduling optimization system, based on the obtained dynamic fit and the process equipment allocation relationship of the preliminary cross-scheme, first determines the current allocated equipment and the optimal fit equipment for each candidate process. Then, it executes equipment replacement according to the principle of prioritizing utilization gap filling and maximizing mold change savings, generating a transitional scheme after equipment allocation variation. The core of its equipment replacement is to improve the overall scheduling scheme's target value by optimizing process-equipment allocation while meeting production constraints. Specifically, in determining the optimal equipment, for each candidate process j, dynamic fit equipment is selected from its set of matching equipment. The equipment with the highest adaptability is selected as the optimal equipment. If multiple equipment have the same adaptability, the equipment with the shortest remaining idle time (≥ total theoretical working time of the process) and shortest mold changeover time is selected. The equipment replacement criteria then include criterion 1 (utilization gap filling): if the utilization gap value of the optimal equipment is... If the dynamic adaptability of the currently assigned equipment is ≥0.1 lower than that of the optimal equipment, then equipment replacement is performed, transferring the candidate process from the currently assigned equipment to the optimal equipment; Criterion 2 (Maximizing mold change savings): If the mold change time of the optimal equipment is shorter than that of the currently assigned equipment, and the potential for savings in cross-equipment mold change is... If the time limit is reached, equipment replacement will be performed. Constraint verification: After replacement, it must be ensured that the theoretical total working time of the candidate process is less than or equal to the remaining idle time of the optimal equipment, and the load rate of the optimal equipment after replacement is less than or equal to 0.95 (to avoid overload). Finally, after the equipment replacement is completed, the actual production time, remaining idle time, cumulative mold change times and other parameters of each equipment are updated, the process priority coding string structure of the original scheme is retained, and only the equipment allocation relationship is adjusted to form a transition scheme.

[0078] Step 40534: Based on the transition scheme and dynamic adaptability, determine the change in mold change time and the improvement in equipment utilization of the mutated process on the newly allocated equipment, and calibrate the priority coding of the mutated process based on the change in mold change time and the improvement in equipment utilization to obtain the mutated scheme.

[0079] Optionally, the production scheduling optimization system, based on the obtained transition plan and dynamic adaptability, first calculates the change in mold changeover time and the increase in equipment utilization for the mutated process (the process after equipment replacement) on the newly allocated equipment. Then, it adjusts the priority coding of the mutated process based on the change in mold changeover time and the increase in equipment utilization, ultimately obtaining the mutated plan. The core of this priority coding calibration is to match the coding weight with the optimization effect after equipment replacement, strengthening the priority of allocating high-quality equipment. Specifically, in calculating the change in mold changeover time... At that time, Refers to the variation process The difference between the actual mold-changing time on the newly allocated equipment and the mold-changing time on the original allocated equipment is calculated using the following formula: ;in, This is the actual mold change time of the newly allocated equipment after the mutation (0 if it is continuously produced with other processes on the same equipment, and the mold change time of the equipment if it is produced independently). This refers to the mold changing time of the originally allocated equipment; This indicates a reduction in mold change time, signifying a positive optimization effect. The increase in equipment utilization is also noted. This refers to the difference between the utilization rate of newly allocated equipment after the change and the utilization rate before the change. The formula is: Among them, the utilization rate of newly allocated equipment after the variation. ( The actual production time of the newly allocated equipment after the mutation; (Total actual available time of newly allocated equipment); Equipment utilization rate of newly allocated equipment before the variation. ( (The actual production time of the newly allocated equipment before the mutation). This indicates improved equipment utilization and a positive optimization effect. Finally, a nonlinear calibration formula that integrates the savings from mold replacement with the increased utilization value is used to intelligently adjust the gene locus encoding corresponding to the mutation process, ensuring a strong correlation between encoding weights and optimization effects. The nonlinear calibration formula is as follows: ; in, For the calibration variation process Priority encoding; For the variation process in the preliminary crossover scheme The original priority encoding; For variation process The maximum value of mold change time for all compatible devices (to ensure that the mold change saving ratio is normalized). Increase the weighting factor for utilization rate (default is 2.5, which strengthens the priority weighting of equipment utilization rate optimization). For variation process The corresponding work order-equipment combination weight lower limit (Level 1) class class ; The boundary constraint function ensures that the range of coded values ​​after calibration is [value missing]. To comply with the coding rules, the updated device allocation relationship and the calibrated priority coding string are integrated after priority coding calibration to form the final mutation scheme. This scheme also needs to re-verify rigid constraints (device matching, time window) and flexible constraints (weight lower limit) to ensure a balance between compliance and optimization.

[0080] This invention accurately identifies processes with optimization potential by constructing a candidate process pool for mutations, avoiding meaningless mutation operations and improving mutation efficiency. Furthermore, a dynamic adaptation model is set up to comprehensively consider multiple dimensions such as equipment idle time, mold change frequency, and load rate, ensuring the scientific and rational nature of equipment replacement. Finally, by closely focusing on the core objectives of maximizing equipment utilization and minimizing mold change time through equipment replacement criteria and priority coding calibration logic, targeted optimization of mutation operations is achieved. This results in a mutation scheme that, while maintaining feasibility, significantly improves the achievement of the core optimization objectives.

[0081] In one embodiment, the process of steps 40541-40544 includes: Step 40541: Based on the variation scheme, the theoretical total working time of the process and the actual available time, determine the process equipment association data for each process; the process equipment association data includes the actual start time, end time, equipment occupancy status and mold change time.

[0082] Optionally, the production scheduling optimization system, based on the obtained variation schemes, the theoretical total working time of each process, and the actual available time, determines the process-equipment association data for each process in each variation scheme through a logic of priority sorting, time window allocation, and resource utilization calculation. This process-equipment association data reflects the adaptation and execution status of the process and equipment, including the actual start time, end time, equipment occupancy, and mold changeover time. Specifically, when calculating the actual start time, the actual start time... Schedule plan intermediate process In the equipment The actual startup time on the device, and meets the requirements of the equipment. The time at which the schedule is available and no earlier than the start time of the scheduling cycle; when calculating the actual end time, the actual end time is... Process In the equipment The actual completion time is calculated using the following formula: ;in, For process The theoretical total working hours for the process; For process In the equipment The mold change time; for mold change time Process In the equipment If the mold change time is continuous with the previous process and the work order type is compatible, then Otherwise, the preset mold change duration is used. Then, in the calculation of the process equipment association data, the execution order of each "work order-equipment" combination is first determined according to the priority encoding string weight of the mutation scheme from high to low. Then, for each process, the earliest available time period is allocated as the actual start time within the idle time window of equipment i to ensure no equipment occupation conflicts. Finally, the actual start time, end time, corresponding equipment occupation time period, and mold change duration of each process are associated and stored to form a process equipment association data matrix.

[0083] Step 40542: Based on the process equipment association data, verify the rigid constraints, repair and re-verify the solutions that violate the rigid constraints, until an intermediate solution that satisfies the constraints and meets the main objective is obtained.

[0084] Optionally, the production scheduling optimization system performs rigid constraint (i.e., equipment matching constraint and time window constraint) checks on the obtained process equipment association data, and performs targeted repairs on solutions that violate the constraints, until an intermediate solution that satisfies the constraints and meets the main objective is generated, thus determining the feasibility of the solution. Specifically, the rigid constraint verification rules include the following for equipment matching constraint verification: the assigned equipment i for process j must belong to the set of compatible equipment for work order j. Otherwise, it is judged as a constraint violation; for time window constraint verification: ① actual end time Production delivery time of work order j ② Actual start time ≥Schedule start time; ③Actual end time The actual available time of device i is ≤ the deadline; any violation of this condition is considered a constraint violation.

[0085] Furthermore, in the constraint repair strategy, for equipment matching constraint repair: process j is reassigned to its matching equipment set. For the equipment with the highest dynamic adaptability, the process equipment association data is recalculated. Regarding time window constraint repair, for overdue delivery repair: the process execution order is adjusted first (reordered according to priority weight), and the idle intervals of non-critical processes are compressed; if overdue delivery still occurs, the process is split into multiple idle time periods on the adapted equipment (process continuity must be met, and the total duration remains unchanged after splitting); for time out-of-bounds repair: the process start time is adjusted to the earliest idle point within the equipment's available time period, and the execution time of subsequent related processes is simultaneously postponed. In the main objective qualification judgment, the repaired solution must meet the main objective value ≥ 0.5 (the comprehensive optimization value of equipment utilization meets the standard to avoid performance degradation after repair); otherwise, the repair process is re-executed. Finally, the solution that passes the constraint verification and meets the main objective is determined as the intermediate solution, and its process equipment association data and target value are retained.

[0086] Step 40543: Merge the elite scheme set and intermediate schemes to obtain the iterative population, and calculate the convergence degree of the primary objective and the convergence degree of the secondary objective based on the iterative population until the convergence degree of the primary objective and the convergence degree of the secondary objective of the iterative population reach the global convergence condition, and obtain the converged target population.

[0087] Optionally, the production scheduling optimization system merges the obtained elite solution set and intermediate solutions to obtain the iterative population for the current round. Then, based on the determined hierarchical objective function, it calculates the primary objective value (equipment utilization rate) and secondary objective value (total mold change time, high-priority work order delivery rate) for each solution in the iterative population. Then, it calculates the convergence of the primary objective and the convergence of the secondary objective respectively. The convergence of the primary objective refers to the absolute value of the difference between the average value of the primary objective value of the current iterative population and the average value of the primary objective value of the previous round iterative population; the convergence of the secondary objective refers to the sum of the absolute values ​​of the differences between the average value of each secondary objective value of the current iterative population and the average value of the corresponding secondary objective value of the previous round. The global convergence condition is set as follows: the convergence of the primary objective is ≤0.5% and the convergence of the secondary objective is ≤1%. The iterative population construction and convergence calculation process is repeated. If the global convergence condition is met, the converged target population is obtained. If the convergence condition is not met, the process returns to step 4051 to re-execute the crossover and mutation operations, update the iterative population, and recalculate the target convergence and secondary objective convergence until the global convergence condition is met, and the converged target population is obtained.

[0088] Step 40544: Fitness evaluation calculation is performed based on the target population and hierarchical objective function to obtain the primary objective value and secondary objective sequence of each scheme. Scheduling scheme is then screened based on the primary objective value and secondary objective sequence to obtain a global scheduling scheme that maximizes equipment utilization, satisfies rigid constraints, and optimizes secondary objectives.

[0089] Optionally, the production scheduling optimization system, based on the determined target population, calls the pre-constructed hierarchical objective function to perform fitness evaluation. First, the primary objective value (equipment utilization) of each scheme is calculated, and the schemes are sorted in descending order of primary objective value. For schemes with a primary objective value difference of less than 1%, the secondary objective sequence (total mold changeover time, high-priority work order delivery rate) is calculated, and the comprehensive value of the secondary objectives is calculated according to the secondary objective hierarchical function and sorted in descending order. According to the preset screening logic: the scheme with the largest primary objective value is selected first; if the primary objective values ​​are the same, the scheme with the largest comprehensive value of the secondary objectives is selected. Finally, the scheme that maximizes equipment utilization, satisfies all rigid constraints, and has the optimal secondary objectives is selected as the optimal global scheduling scheme.

[0090] This invention provides a precise basis for constraint verification and fitness evaluation by refining the associated data of process equipment. The verification-repair closed loop of rigid constraints ensures the feasibility of the solution and avoids production conflicts. Furthermore, the iterative termination logic based on convergence ensures that the algorithm converges to a stable optimization region, avoiding getting trapped in local optima. Finally, fitness evaluation and screening guided by a hierarchical objective function ultimately locks in a global scheduling scheme that maximizes equipment utilization and achieves suboptimal secondary objectives. This comprehensively solves the core pain points of constraint compliance, multi-objective balance, and optimization stability in mold scheduling, providing an efficient and reliable optimal solution for production execution.

[0091] Furthermore, the production scheduling optimization system based on genetic algorithm provided by the present invention will be described below. The production scheduling optimization system based on genetic algorithm described below can be referred to in correspondence with the production scheduling optimization method based on genetic algorithm described above.

[0092] Optional, refer to Figure 2 , Figure 2 This is a schematic diagram of the production scheduling optimization system based on genetic algorithm provided by the present invention. The production scheduling optimization system based on genetic algorithm includes: Data acquisition module 210 is used to acquire work order data, mold process data and resource data required in the production process based on ERP system and MES system; The calculation and processing module 220 is used to determine the theoretical total working hours of each process corresponding to each work order based on work order data and mold process data, and to determine the actual available time of each production equipment in each future shift based on resource data combined with the production resource calendar. The initial scheme generation module 230 is used to construct coding rules based on work order data, mold process data and resource data, with the process priority sequence as the core, and to perform population initialization based on the coding rules and the pre-built decoder to obtain the initial scheduling scheme. The scheme iteration optimization module 240 is used to perform fitness evaluation and evolution iteration based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, so as to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The scheduling visualization and execution module 250 is used to perform graphical transformation based on the optimal global scheduling scheme to obtain a scheduling Gantt chart, and enable planners to execute production operations according to the scheduling Gantt chart.

[0093] This invention, through acquiring work order data, mold process data, and resource data, determines the theoretical total working time of each process corresponding to each work order and the actual available time of each production equipment in each future shift. It clarifies the time benchmark for work order execution and the capacity boundary of the equipment. Based on this data, it constructs an encoding rule centered on the process priority sequence and combines it with a pre-built decoder to complete population initialization, obtaining an initial scheduling scheme that meets the basic production constraints. This ensures that the quality of the initial population is higher than that of random solutions, thereby reducing the search area for a large amount of mold change costs, improving the convergence speed and the quality of the final solution. Then, based on the initial scheduling scheme, the theoretical total working time of the processes, and the actual available time of the equipment, it achieves global optimization of work order sorting and equipment allocation through fitness evaluation and evolutionary iteration. This effectively balances multiple objectives such as work order delivery, equipment utilization, and mold change costs, while strengthening the ability to handle complex production constraints. Finally, the optimal global scheduling scheme is transformed into an intuitive scheduling Gantt chart, facilitating efficient production operations by planners. In summary, this invention effectively solves the technical pain points of existing scheduling methods, such as weak constraint processing capability, high mold change cost, difficulty in balancing multiple objectives, and slow dynamic response, thereby improving the scientific nature and execution efficiency of scheduling schemes.

[0094] Please see Figure 3 , Figure 3 An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, this embodiment of the invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following steps: Based on the ERP system and MES system, obtain the work order data, mold process data and resource data required in the production process; Based on work order data and mold process data, the theoretical total working hours of each process corresponding to each work order are determined, and based on resource data combined with the production resource calendar, the actual available time of each production equipment in each future shift is determined. Based on work order data, mold process data, and resource data, a coding rule with process priority sequence as the core is constructed, and population initialization is performed based on the coding rule and pre-built decoder to obtain an initial scheduling scheme. Based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, the fitness evaluation and evolutionary iteration are carried out to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The optimal global scheduling scheme is used to perform graphical transformation to obtain a scheduling Gantt chart, which allows planners to execute production operations based on the scheduling Gantt chart.

[0095] Please see Figure 4 , Figure 4An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it performs the following steps: Based on the ERP system and MES system, obtain the work order data, mold process data and resource data required in the production process; Based on work order data and mold process data, the theoretical total working hours of each process corresponding to each work order are determined, and based on resource data combined with the production resource calendar, the actual available time of each production equipment in each future shift is determined. Based on work order data, mold process data, and resource data, a coding rule with process priority sequence as the core is constructed, and population initialization is performed based on the coding rule and pre-built decoder to obtain an initial scheduling scheme. Based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, the fitness evaluation and evolutionary iteration are carried out to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The optimal global scheduling scheme is used to perform graphical transformation to obtain a scheduling Gantt chart, which allows planners to execute production operations based on the scheduling Gantt chart.

[0096] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the production scheduling optimization method based on genetic algorithms provided by the above methods, the method including: Based on the ERP system and MES system, obtain the work order data, mold process data and resource data required in the production process; Based on work order data and mold process data, the theoretical total working hours of each process corresponding to each work order are determined, and based on resource data combined with the production resource calendar, the actual available time of each production equipment in each future shift is determined. Based on work order data, mold process data, and resource data, a coding rule with process priority sequence as the core is constructed, and population initialization is performed based on the coding rule and pre-built decoder to obtain an initial scheduling scheme. Based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, the fitness evaluation and evolutionary iteration are carried out to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The optimal global scheduling scheme is used to perform graphical transformation to obtain a scheduling Gantt chart, which allows planners to execute production operations based on the scheduling Gantt chart.

[0097] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0098] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A production scheduling optimization method based on genetic algorithm, characterized by, include: Based on the ERP system and MES system, obtain the work order data, mold process data and resource data required in the production process; Based on the work order data and the mold process data, the theoretical total working hours of each process corresponding to each work order are determined, and based on the resource data and the production resource calendar, the actual available time of each production equipment in each future shift is determined. Based on the work order data, mold process data, and resource data, a coding rule centered on the process priority sequence is constructed. Population initialization is then performed based on this coding rule and a pre-built decoder to obtain an initial scheduling scheme. The work order data includes work order number, product code, required quantity, production delivery date, and priority. The mold process data includes mold group, number of mold cavities, standard mold closing and changeover time, pass rate, material and color characteristics. The resource data includes mold equipment data and production resource data. The production resource data includes equipment status, shift plan, exceptional downtime, and real-time efficiency. The method for constructing coding rules based on the work order data, the mold process data, and the resource data, with the process priority sequence as the core, includes: Based on the product code and required quantity in the work order data, and the mold group, number of mold cavities, and pass rate in the mold process data, the basic parameters for process adaptation between the work order and the mold group are determined, and the process adaptation index corresponding to each work order is determined based on the basic parameters for process adaptation; the process adaptation index characterizes the degree of process matching between the work order and the corresponding mold group. Based on the process adaptability index, combined with the production delivery date and priority in the work order data, and the equipment status and shift plan in the production resource data, the adaptability correlation parameters between the work order and the production resources are determined, and the resource availability index corresponding to each work order is determined based on the adaptability correlation parameters; the resource availability index characterizes the degree of immediate availability matching of the production resources corresponding to the work order. Based on the process adaptability index and the resource availability index, a basic priority coefficient for each work order is constructed. Based on the basic priority coefficient and the material and color characteristics and standard mold change time in the mold process data, the mold change association parameters of the work order are determined, and the mold change adaptation correction coefficient of each work order is determined based on the mold change association parameters; the mold change adaptation correction coefficient is used to correct the mold change process influence not considered in the basic priority coefficient. Based on the basic priority coefficient and the mold change adaptation correction coefficient, the target priority value of each work order is calculated, and the work orders are sorted from largest to smallest according to the target priority value to obtain the coding rule with the process priority sequence as the core; wherein, the arrangement order of work orders in the sequence corresponds to the arrangement order of chromosome genes, and each work order corresponds to a gene identifier. Based on the initial scheduling scheme, the theoretical total working time of the process, and the actual available time, fitness evaluation and evolutionary iteration are performed to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. Based on the optimal global scheduling scheme, a graphical transformation is performed to obtain a scheduling Gantt chart, and the planner executes production operations according to the scheduling Gantt chart.

2. The genetic algorithm-based production scheduling optimization method according to claim 1, characterized by, The process of performing fitness evaluation and evolutionary iteration based on the initial scheduling scheme, the theoretical total working time of the process, and the actual available time to obtain an optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints includes: Based on the theoretical total working hours of the process, the actual available time, and the initial scheduling scheme, constraint mapping is performed to obtain rigid constraints and flexible constraints for each process. Based on the rigid constraints and flexible constraints, a hierarchical objective function combining primary and secondary objective requirements is constructed. The rigid constraints include time window constraints and equipment matching constraints. The flexible constraints are priority constraints. Based on the hierarchical objective function, the primary objective value and secondary objective value of each scheme in the initial scheduling scheme are calculated to obtain the set of scheme objective values; Based on the set of target values ​​of the schemes, the schemes are sorted in descending order according to the primary target value. Schemes with a difference of less than a preset value in the primary target value are divided into the same primary target group. The schemes in the same primary target group are sorted in non-dominated order according to the secondary target value to obtain the secondary target non-dominated level. The secondary target crowding distance of the schemes in the same primary target group is calculated to obtain the fitness evaluation result of each scheme. Based on the fitness evaluation results, the initial scheduling scheme is screened and retained to obtain an elite scheme set. The remaining schemes in the initial scheduling scheme are then screened using the roulette wheel selection method according to the order of primary objective grouping priority and secondary objective crowding distance to obtain a set of schemes to be crossovered. Based on the elite scheme set and the scheme set to be crossed, the population is cross-crossed, mutated, and iteratively terminated to obtain the optimal global scheduling scheme.

3. The genetic algorithm-based production scheduling optimization method according to claim 2, characterized by, The process of performing crossover, mutation, and iterative termination on the population based on the elite scheme set and the scheme set to be crossovered, to obtain the optimal global scheduling scheme, includes: Based on the set of cross-scheduling schemes to be crossed and the coding rules of the initial scheduling scheme, the process priority coding string of the scheme is determined; Based on the process priority encoding string, the process priority segments between any two intersection points in the set of crossover schemes are swapped, and the process priority order and equipment-process matching correlation outside the intersection point are preserved to obtain a preliminary crossover scheme. Based on the preliminary crossover scheme and the utilization rate and idle time information of each device determined by the actual available time, a variation operation is performed in combination with the theoretical total working time of the process to obtain the variation scheme; Based on the elite solution set, the variant solution, the theoretical total working time of the process, and the actual available time, the optimal global scheduling solution is obtained through iterative updates and solution selection.

4. The genetic algorithm-based production scheduling optimization method according to claim 2, characterized by, The primary objective requirement is equipment utilization rate; the secondary objective requirements include total mold changeover time, on-time order delivery rate, and work order priority. The hierarchical objective function, constructed based on the rigid and flexible constraints and combining primary and secondary objective requirements, includes: Based on the feasibility constraints in the mold scheduling scenario, a feasibility verification layer is constructed with the constraint compliance judgment rules of the scheduling scheme as the core. The initial scheduling scheme is then screened within the feasibility verification layer to obtain a set of feasible scheduling schemes. Based on the set of feasible scheduling schemes, combined with the actual available time of the equipment and the theoretical total working time of the process, the actual production time of the equipment corresponding to each feasible scheduling scheme is determined. Based on the actual production time of the equipment, a core target layer is constructed with equipment utilization rate as the core. In the core target layer, the feasible scheduling schemes are sorted and filtered to obtain a subset of feasible scheduling schemes with similar equipment utilization rates after sorting. Based on the subset of feasible scheduling schemes, combined with work order priority, on-time delivery rate and mold changeover time, the total mold changeover time and high-priority work order delivery status corresponding to each scheme are determined, and a secondary target layer is constructed based on the total mold changeover time and high-priority work order delivery status. The hierarchical objective function is obtained by integrating the feasibility verification layer, the core objective layer, and the secondary objective layer.

5. The genetic algorithm-based production scheduling optimization method according to claim 3, characterized by, The process of iteratively updating and filtering solutions based on the elite solution set, the variant solutions, the theoretical total working time of the process, and the actual available time to obtain the optimal global scheduling solution includes: Based on the aforementioned variation scheme, the theoretical total working time of the process, and the actual available time, the process equipment association data for each process is determined; the process equipment association data includes the actual start time, end time, equipment occupancy status, and mold change time; Based on the process equipment association data, the rigid constraints are verified, and the solutions that violate the rigid constraints are repaired and re-verified until an intermediate solution that satisfies the constraints and meets the main objective is obtained. The elite scheme set and the intermediate scheme are merged to obtain an iterative population. The convergence of the primary objective and the secondary objective is calculated based on the iterative population until the convergence of the primary objective and the secondary objective of the iterative population reaches the global convergence condition, thus obtaining the converged target population. Fitness evaluation calculations are performed based on the target population and the hierarchical objective function to obtain the primary objective value and secondary objective sequence for each scheme. Scheduling schemes are then selected based on the primary objective value and the secondary objective sequence to obtain the global scheduling scheme that maximizes equipment utilization, satisfies rigid constraints, and optimizes the secondary objectives.

6. The production scheduling optimization method based on genetic algorithm according to claim 3, characterized in that, The utilization rate and idle time information of each device determined based on the preliminary crossover scheme and actual available time are combined with the theoretical total working time of the process to perform a variation operation to obtain a variation scheme, including: Based on the preliminary cross-process scheme, the theoretical total working time of the process, and the actual available time, the utilization gap value of each piece of equipment and the cross-equipment mold change saving potential of each process are determined. Based on the utilization gap value and the cross-equipment mold change saving potential, a pool of variant candidate processes is constructed. The cross-equipment mold change saving potential refers to the difference in mold change time between different pieces of equipment. Based on the remaining idle time, cumulative mold change count, and current load rate of each matching device within the time window of each candidate process determined by the mutated candidate process pool and actual available time, the dynamic adaptability of the device to the candidate process is constructed. Based on the dynamic adaptability and the process equipment allocation relationship of the preliminary cross scheme, the current allocated equipment and the optimal adaptability equipment for each candidate process are determined. Based on the current allocated equipment and the optimal equipment, equipment replacement is carried out according to the criteria of filling the utilization gap and saving mold replacement, so as to obtain the transition scheme after the equipment allocation variation. Based on the aforementioned transition scheme and dynamic adaptability, the change in mold-changing time and the increase in equipment utilization rate of the mutated process on the newly allocated equipment are determined. Based on the change in mold-changing time and the increase in equipment utilization rate, the priority coding of the mutated process is calibrated to obtain the mutated scheme.

7. A production scheduling optimization system based on a genetic algorithm, characterized in that, Applied to the production scheduling optimization method based on genetic algorithms as described in any one of claims 1 to 6; The production scheduling optimization system based on genetic algorithms includes: The data acquisition module is used to acquire work order data, mold process data, and resource data required during the production process based on the ERP system and MES system; The calculation and processing module is used to determine the theoretical total working hours of each process corresponding to each work order based on the work order data and the mold process data, and to determine the actual available time of each production equipment in each future shift based on the resource data and the production resource calendar. The initial scheme generation module is used to construct coding rules based on the work order data, the mold process data, and the resource data, with the process priority sequence as the core, and to perform population initialization based on the coding rules and a pre-built decoder to obtain an initial scheduling scheme. The work order data includes work order number, product code, required quantity, production delivery date, and priority. The mold process data includes mold group, number of mold cavities, standard mold closing and mold changing time, pass rate, material and color characteristics. The resource data includes mold equipment data and production resource data. The production resource data includes equipment status, shift plan, exceptional downtime, and real-time efficiency. The method for constructing coding rules based on the work order data, the mold process data, and the resource data, with the process priority sequence as the core, includes: Based on the product code and required quantity in the work order data, and the mold group, number of mold cavities, and pass rate in the mold process data, the basic parameters for process adaptation between the work order and the mold group are determined, and the process adaptation index corresponding to each work order is determined based on the basic parameters for process adaptation; the process adaptation index characterizes the degree of process matching between the work order and the corresponding mold group. Based on the process adaptability index, combined with the production delivery date and priority in the work order data, and the equipment status and shift plan in the production resource data, the adaptability correlation parameters between the work order and the production resources are determined, and the resource availability index corresponding to each work order is determined based on the adaptability correlation parameters; the resource availability index characterizes the degree of immediate availability matching of the production resources corresponding to the work order. Based on the process adaptability index and the resource availability index, a basic priority coefficient for each work order is constructed. Based on the basic priority coefficient and the material and color characteristics and standard mold change time in the mold process data, the mold change association parameters of the work order are determined, and the mold change adaptation correction coefficient of each work order is determined based on the mold change association parameters; the mold change adaptation correction coefficient is used to correct the mold change process influence not considered in the basic priority coefficient. Based on the basic priority coefficient and the mold change adaptation correction coefficient, the target priority value of each work order is calculated, and the work orders are sorted from largest to smallest according to the target priority value to obtain the coding rule with the process priority sequence as the core; wherein, the arrangement order of work orders in the sequence corresponds to the arrangement order of chromosome genes, and each work order corresponds to a gene identifier. The scheme iteration optimization module is used to perform fitness evaluation and evolution iteration based on the initial scheduling scheme, the theoretical total working time of the process and the actual available time, so as to obtain the optimal global scheduling scheme with high equipment utilization and satisfying multiple constraints. The scheduling visualization and execution module is used to perform graphical transformation based on the optimal global scheduling scheme to obtain a scheduling Gantt chart, and to enable planners to execute production operations based on the scheduling Gantt chart.

8. An electronic device, characterized in that, include: Memory, used to store computer software programs; A processor is configured to read and execute the computer software program, wherein when the processor executes the computer software program, it implements the production scheduling optimization method based on genetic algorithm as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, characterized in that, The storage medium stores a computer software program, which, when executed by a processor, implements the production scheduling optimization method based on a genetic algorithm as described in any one of claims 1 to 6.

Images