Production scheme optimization method and device, equipment and medium

Abstract

The invention belongs to the field of production and manufacturing optimization, and particularly relates to a production scheme optimization method and device, equipment and a medium, and the method comprises the steps: generating a filial generation production scheme through the crossover and mutation operation of a genetic algorithm, so as to increase the diversity; dividing the combined candidate schemes into non-dominated layers with different quality levels according to the dominated relationship of the plurality of cost parameters; the production schemes are selected according to the priorities of the non-dominated layers from high to low, the distributivity of a solution set is guaranteed through crowding degree calculation in the same layer, the population is made to continuously approach the real optimal leading edge through iterative evolution, and finally the optimal production scheme is output. According to the method, under the condition that multiple targets are not artificially weighted into a single target, a batch of optimization schemes capable of achieving good balance among multiple competitive targets can be automatically found. Compared with a traditional optimization method, the method not only can ensure that the quality of the solution is quickly converged to the Pareto frontier, but also can maintain the diversity of the solution set through a crowding degree mechanism, and effectively avoids falling into local optimum.

Description

Production scheme optimization methods, devices, equipment and media Technical Field

[0001] This application belongs to the field of production optimization, specifically relating to a method, apparatus, equipment and medium for optimizing production schemes. Background Technology

[0002] Currently, complex manufacturing scenarios under multiple constraints are emerging, such as personalized customization, resource-constrained flexible production lines, and dynamically changing production environments. These scenarios need to simultaneously meet customers' multi-dimensional needs for product configuration, performance, delivery time, and cost, achieve efficient scheduling with limited resources, and be able to cope with emergencies such as order changes and equipment failures.

[0003] Existing technologies mostly employ traditional single-objective optimization methods, such as linear programming or dynamic programming, which optimize only around a single objective (such as minimizing time or cost). These methods struggle to achieve an effective balance among multiple objectives and have limited ability to handle complex constraints (such as resource conflicts and process dependencies), making them unable to adapt to dynamically changing production environments and hindering the improvement of the overall performance of manufacturing systems.

[0004] Therefore, in order to solve the problems of traditional manufacturing models that can only provide a single service and have low resource utilization efficiency, it is urgent to build a production optimization method oriented towards multi-objective optimization. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a method, apparatus, equipment, and medium for optimizing production plans.

[0006] Firstly, this application provides a method for optimizing a production scheme, the method comprising:

[0007] Obtain a set of parent production plans, which includes multiple production plans for the same production task, and each production plan includes multiple cost parameters;

[0008] Perform crossover and mutation operations on the parent production scheme set to obtain the child production scheme set;

[0009] The parent production scheme set and the child production scheme set are merged to obtain a candidate production scheme set. Based on the multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set, the candidate production schemes are sorted in a non-dominated hierarchical manner to obtain a multi-level non-dominated set. The priority of the non-dominated set decreases sequentially as the number of levels increases.

[0010] The first N production schemes of the multi-level non-dominated set are combined into a new parent production scheme set. Based on the new parent production scheme set, the steps of the crossover and mutation operations are returned for iterative processing until the iteration termination condition is met, and the final parent production scheme is obtained.

[0011] In one possible implementation, the step of performing a non-dominated hierarchical sorting of the candidate production schemes based on multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set includes:

[0012] For each candidate production scheme in the candidate production scheme set, based on multiple cost parameters of each candidate production scheme, determine the number of winning production schemes and the set of losing production schemes corresponding to each candidate production scheme. The number of winning production schemes corresponding to each candidate production scheme is the number of winning production schemes whose cost parameters are all better than the corresponding candidate production scheme, and the cost parameters of each candidate production scheme are all better than each losing production scheme in the corresponding set of losing production schemes.

[0013] Based on the number of winning production schemes corresponding to each candidate production scheme and the set of losing production schemes, the candidate production schemes in the candidate production scheme set are sorted in a non-dominated hierarchical manner to obtain a multi-level non-dominated set.

[0014] In one possible implementation, the step of performing a non-dominated hierarchical sorting of the candidate production schemes in the candidate production scheme set based on the number of winning production schemes corresponding to each candidate production scheme and the set of losing production schemes, to obtain a multi-level non-dominated set, includes:

[0015] For the first layering operation, the candidate production schemes with zero winning production schemes are determined as the first layer production schemes, and the first layer production schemes are placed into the first layer non-dominated set.

[0016] For subsequent non-first-time stratification operations, based on the set of inferior production schemes corresponding to each production scheme in the previous layer, the latest number of superior production schemes for each inferior production scheme in the set of inferior production schemes is reduced by one to obtain the updated number of superior production schemes.

[0017] The inferior production scheme with zero winning production schemes after the update is identified as the current layer production scheme, and the current layer production scheme is placed into the current layer non-dominated set, until all candidate production schemes in the candidate production scheme set are placed into their corresponding non-dominated sets.

[0018] In one possible implementation, combining the first N production schemes of the plurality of non-dominated sets into a new parent production scheme set includes:

[0019] Starting from the first layer of non-dominated sets, the full production plans from each layer of non-dominated sets are added to the new parent production plan set in sequence.

[0020] During the addition process, it is determined whether the number of production schemes in the new parent production scheme set has reached N;

[0021] If not, continue adding the full production schemes of the next layer of non-dominated sets until the number of production schemes in the new parent production scheme set reaches N.

[0022] In one possible implementation, the method further includes:

[0023] If, when adding the Kth layer non-dominated set, all Kth layer production schemes of the Kth layer non-dominated set are added, the number of production schemes in the new parent production scheme set will exceed N, then the Kth layer production schemes in the Kth layer non-dominated set will be sorted by congestion.

[0024] Based on the crowding ranking, the required number of production schemes for the Kth layer are selected sequentially from the sorted non-dominated set of the Kth layer, so that the number of production schemes in the new parent production scheme set is equal to N.

[0025] In one possible implementation, the congestion ranking of the Kth-level production schemes in the Kth-level non-dominated set includes:

[0026] Based on the value of each cost parameter corresponding to the Kth layer production scheme, an ascending sequence of the Kth layer production schemes corresponding to each cost parameter is obtained;

[0027] For the Mth Kth layer production scheme in a single ascending sequence, calculate the difference in corresponding cost parameters between the (M-1)th Kth layer production scheme and the (M+1)th Kth layer production scheme;

[0028] For each Kth-level production scheme, the differences calculated for each cost parameter are summed to obtain the congestion degree corresponding to each Kth-level production scheme;

[0029] Based on the congestion level corresponding to each of the Kth-level production schemes, the Kth-level production schemes in the Kth-level non-dominated set are sorted by congestion level.

[0030] In one possible implementation, the iteration termination condition includes: the number of iterations reaches a preset number.

[0031] Secondly, this application provides a production scheme optimization device, the device comprising:

[0032] The acquisition module is used to acquire a set of parent production plans, which includes multiple production plans for the same production task, and each production plan includes multiple cost parameters.

[0033] The execution module is used to perform crossover and mutation operations on the parent production scheme set to obtain the child production scheme set;

[0034] The processing module is used to merge the parent production scheme set and the child production scheme set to obtain a candidate production scheme set, and to perform non-dominated hierarchical sorting of the candidate production schemes based on multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set to obtain a multi-level non-dominated set, wherein the priority of the non-dominated set decreases sequentially as the number of levels increases.

[0035] The iteration module is used to combine the first N production schemes of the multi-level non-dominated set into a new set of parent production schemes. Based on the new set of parent production schemes, the steps of the crossover and mutation operations are returned for iterative processing until the iteration termination condition is met, and the final parent production scheme is obtained.

[0036] In one possible implementation, the processing module is specifically used for:

[0037] For each candidate production scheme in the candidate production scheme set, based on multiple cost parameters of each candidate production scheme, determine the number of winning production schemes and the set of losing production schemes corresponding to each candidate production scheme. The number of winning production schemes corresponding to each candidate production scheme is the number of winning production schemes whose cost parameters are all better than the corresponding candidate production scheme, and the cost parameters of each candidate production scheme are all better than each losing production scheme in the corresponding set of losing production schemes.

[0038] Based on the number of winning production schemes corresponding to each candidate production scheme and the set of losing production schemes, the candidate production schemes in the candidate production scheme set are sorted in a non-dominated hierarchical manner to obtain a multi-level non-dominated set.

[0039] In one possible implementation, the processing module is specifically used for:

[0040] For the first layering operation, the candidate production schemes with zero winning production schemes are determined as the first layer production schemes, and the first layer production schemes are placed into the first layer non-dominated set.

[0041] For subsequent non-first-time stratification operations, based on the set of inferior production schemes corresponding to each production scheme in the previous layer, the latest number of superior production schemes for each inferior production scheme in the set of inferior production schemes is reduced by one to obtain the updated number of superior production schemes.

[0042] The inferior production scheme with zero winning production schemes after the update is identified as the current layer production scheme, and the current layer production scheme is placed into the current layer non-dominated set, until all candidate production schemes in the candidate production scheme set are placed into their corresponding non-dominated sets.

[0043] In one possible implementation, the iteration module is specifically used for:

[0044] Starting from the first layer of non-dominated sets, the full production plans from each layer of non-dominated sets are added to the new parent production plan set in sequence.

[0045] During the addition process, it is determined whether the number of production schemes in the new parent production scheme set has reached N;

[0046] If not, continue adding the full production schemes of the next layer of non-dominated sets until the number of production schemes in the new parent production scheme set reaches N.

[0047] In one possible implementation, the iteration module is also used for:

[0048] If, when adding the Kth layer non-dominated set, all Kth layer production schemes of the Kth layer non-dominated set are added, the number of production schemes in the new parent production scheme set will exceed N, then the Kth layer production schemes in the Kth layer non-dominated set will be sorted by congestion.

[0049] Based on the crowding ranking, the required number of production schemes for the Kth layer are selected sequentially from the sorted non-dominated set of the Kth layer, so that the number of production schemes in the new parent production scheme set is equal to N.

[0050] In one possible implementation, the iteration module is specifically used for:

[0051] Based on the value of each cost parameter corresponding to the Kth layer production scheme, an ascending sequence of the Kth layer production schemes corresponding to each cost parameter is obtained;

[0052] For the Mth Kth layer production scheme in a single ascending sequence, calculate the difference in corresponding cost parameters between the (M-1)th Kth layer production scheme and the (M+1)th Kth layer production scheme;

[0053] For each Kth-level production scheme, the differences calculated for each cost parameter are summed to obtain the congestion degree corresponding to each Kth-level production scheme;

[0054] Based on the congestion level corresponding to each of the Kth-level production schemes, the Kth-level production schemes in the Kth-level non-dominated set are sorted by congestion level.

[0055] In one possible implementation, the iteration module includes an iteration termination condition that includes: the number of iterations reaches a preset number.

[0056] Thirdly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of the first aspects.

[0057] Fourthly, this application provides an electronic device, comprising: at least one processor and a memory; wherein,

[0058] The memory stores computer-executed instructions;

[0059] The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method as described in any of the first aspects.

[0060] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, can implement the steps of the method as described in any of the first aspects.

[0061] The production scheme optimization method, apparatus, equipment, and medium provided in this application generate offspring schemes to increase diversity through crossover and mutation operations in a genetic algorithm. Then, using non-dominated sorting, based on the dominance relationships of multiple cost parameters (such as time and cost), the merged candidate schemes are divided into non-dominated layers of varying quality. Finally, schemes are selected from high to low priority according to the non-dominated layer, and the distribution of the solution set is ensured within the same layer through crowding calculation. Iterative evolution allows the population to continuously approach the true optimal frontier, ultimately outputting the optimal production scheme. This method can automatically find a batch of optimized schemes that achieve a good balance among multiple competing objectives without artificially weighting multiple objectives into a single objective, providing decision-makers with a rich selection space. Compared with traditional optimization methods, it can ensure the quality of the solution converges quickly to the Pareto front, and maintain the diversity of the solution set through the crowding mechanism, effectively avoiding getting trapped in local optima. It is particularly suitable for solving multi-objective optimization problems with conflicting objectives in complex production environments. Attached Figure Description

[0062] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0063] Figure 1 is a flowchart of the production scheme optimization method provided in an embodiment of this application;

[0064] Figure 2 is a flowchart of the production scheme optimization method provided in the embodiment of this application;

[0065] Figure 3 is a flowchart of the production scheme optimization method provided in the embodiment of this application;

[0066] Figure 4 is a production process flowchart provided in an embodiment of the present invention;

[0067] Figure 5 shows the distribution of the shortest production time for the reorganized production schemes obtained by three different calculation strategies.

[0068] Figure 6 shows the distribution of the lowest production cost of the reorganized production schemes obtained by three different calculation strategies;

[0069] Figure 7 is a diagram of a production scheme optimization device provided in an embodiment of the present invention;

[0070] Figure 8 is a hardware schematic diagram of the electronic device provided in an embodiment of the present invention.

[0071] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0073] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein.

[0074] In this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0075] In manufacturing systems, developing optimal production strategies is crucial for improving product quality and manufacturing efficiency. Faced with complex processes and fluctuating demands, optimized strategies not only relate to material allocation, process routing, and equipment utilization, but also impact product quality, delivery cycles, and cost control. Efficient production strategies help improve resource utilization, reduce energy consumption and scrap rates, and enhance system flexibility and intelligence, thus playing a vital role in achieving high-quality and sustainable development in the manufacturing industry.

[0076] Currently, customer needs are evolving from single-function product requirements to personalized demands encompassing multiple dimensions such as configuration, performance, quality, delivery time, and cost. Faced with these diversified and dynamic demands, traditional manufacturing models, due to their rigid structure and low responsiveness, struggle to meet highly customized production requirements. Therefore, to address the issues of traditional manufacturing models—such as providing only a single service and low resource utilization efficiency—there is an urgent need to develop production optimization methods oriented towards multi-objective optimization.

[0077] To address the problems in existing technologies, this application provides a production scheme optimization method. It generates offspring schemes to increase diversity through crossover and mutation operations in a genetic algorithm. Then, using non-dominated sorting, based on the dominance relationships of multiple cost parameters (such as time and cost), the merged candidate schemes are divided into non-dominated layers of varying quality. Finally, schemes are selected from high to low priority according to the non-dominated layer, and the distribution of the solution set is ensured within the same layer through crowding calculation. Iterative evolution allows the population to continuously approach the true optimal frontier, ultimately outputting the optimal production scheme. This method can automatically find a batch of optimized schemes that achieve a good balance among multiple competing objectives without artificially weighting multiple objectives into a single objective, providing decision-makers with a rich selection space. Compared with traditional optimization methods, it can both ensure the quality of the solution converges quickly to the Pareto front and maintain the diversity of the solution set through the crowding mechanism, effectively avoiding getting trapped in local optima. It is particularly suitable for solving multi-objective optimization problems with conflicting objectives in complex production environments.

[0078] The technical solutions of this application and how they solve the aforementioned technical problems are described in detail below with specific embodiments. These specific embodiments may exist independently or in combination with each other. Identical or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0079] This embodiment provides a production plan optimization method. Figure 1 is a flowchart of the production plan optimization method provided in this embodiment. The method includes:

[0080] S101. Obtain the parent production plan set. The parent production plan set includes multiple production plans for the same production task, and each production plan includes multiple cost parameters.

[0081] In this step, the parent production plan set is the starting point for algorithm iteration, equivalent to the first generation of the population in the evolutionary process. It contains multiple different, feasible solutions, providing material for subsequent crossover and mutation. A production plan refers to a complete execution plan for a specific production task (such as producing 1000 parts of a certain model). It can be encoded as a unique piece of information, such as a vector or sequence, containing decision variables such as process sequencing, machine allocation, and parameter settings. Cost parameters are multi-dimensional objectives used to evaluate the quality of a production plan. Examples include: production costs (materials, energy consumption, labor), production time (completion cycle), resource utilization, and product quality (pass rate). These parameters constitute the objective space for plan optimization.

[0082] For example, within a solution space that satisfies all production constraints (such as machine capacity and process sequence), a random number generator assigns feasible random values ​​to each decision variable, thereby generating a completely random set of initial solutions. Alternatively, domain knowledge or classic scheduling rules (such as shortest processing time priority or earliest delivery date priority) can be used to systematically construct initial solutions. By embedding human expert experience or mature rules into the initialization process, it is possible to quickly generate a set of benchmark solutions that perform well on a specific objective.

[0083] S102. Perform crossover and mutation operations on the parent production scheme set to obtain the child production scheme set.

[0084] In this step, crossover involves selecting two schemes from the parent generation and exchanging a portion of their genes (i.e., decision variables) to generate new offspring schemes. Its core objective is to utilize existing superior genes to create potentially better new solutions. Mutation involves randomly modifying some decision variables of a single scheme to introduce new genetic traits. Its core objective is to maintain population diversity, helping the algorithm escape local optima and explore new solution spaces.

[0085] For example, when performing crossover operations, if the production plan is represented as a sequence of processes (such as assembly line scheduling), sequential crossover can first randomly select a subsequence from parent generation one, and then traverse from the starting position of parent generation two, filling the subsequence with processes from parent generation one that do not appear in the subsequence in sequence, thereby generating a new sequence that retains some of the structure of the parent generation while incorporating the genes of the other. Alternatively, when the cost parameter is a continuous value (such as temperature or pressure), this method simulates the effect of single-point crossover, using a distribution index to control the degree of deviation between the offspring and the parent generation. The generated offspring parameter values ​​will be closer to the parent generation with a higher probability and farther away from the parent generation with a lower probability, achieving the purpose of fine-grained search around the parent generation.

[0086] For example, when performing mutation operations, if it is sequence-based encoding, it can be achieved by randomly selecting two or more positions in the sequence and swapping their values. For instance, in a machine allocation scheme, randomly swapping the machines used in two processes can generate a new permutation locally. Alternatively, if it is a continuous variable, a random perturbation can be added to the current parameter value. The magnitude of this perturbation is determined by a multinomial distribution, which allows for fine-grained exploration with small steps or larger jumps, effectively maintaining the diversity of the population in the continuous space.

[0087] S103. Merge the parent production scheme set and the child production scheme set to obtain the candidate production scheme set.

[0088] In this step, the candidate production scheme set is a larger temporary population formed by merging the parent and offspring populations. This set forms the basis for elite selection and non-dominated ranking. The parent population and the offspring population generated through crossover mutation can be directly combined to form a merged population that is twice the size. Alternatively, a deduplication step can be added to the simple merging process, identifying and removing identical duplicate individuals by comparing the scheme's encoding or objective function values.

[0089] S104. Based on the multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set, the candidate production schemes are sorted by non-dominated hierarchical order to obtain a multi-level non-dominated set. The priority of the non-dominated set decreases sequentially as the number of levels increases.

[0090] In this step, non-dominated stratification sorting refers to stratifying individuals in the merged population according to multiple objective functions (cost parameters). A non-dominated set refers to a set where no two solutions dominate each other within the same stratum. The first non-dominated stratum contains all individuals not dominated by any other solution and represents the optimal solution set in the current population; the second non-dominated stratum is the optimal solution set among the remaining individuals after removing individuals from the first stratum, and so on. Lower strata have higher priority.

[0091] For example, all non-dominant individuals can be identified from the entire population to form the first layer; then, these individuals are removed from the population, and the above process is repeated among the remaining individuals to identify new non-dominant individuals to form the second layer; this process is repeated recursively until the population is empty.

[0092] For example, based on multiple cost parameters corresponding to each candidate production plan in the candidate production plan set, a non-dominated hierarchical ranking of the candidate production plans is performed, including:

[0093] For each candidate production plan in the candidate production plan set, based on multiple cost parameters of each candidate production plan, determine the number of winning production plans and the set of inferior production plans for each candidate production plan. The number of winning production plans for each candidate production plan is the number of winning production plans whose cost parameters are all better than the corresponding candidate production plan. The cost parameters of each candidate production plan are better than each inferior production plan in the set of inferior production plans.

[0094] Based on the number of winning production schemes corresponding to each candidate production scheme and the set of losing production schemes, the candidate production schemes in the candidate production scheme set are sorted in a non-dominated hierarchical manner to obtain a multi-level non-dominated set.

[0095] In this example, the number of winning production solutions refers to the number of other solutions that are comprehensively superior to the current candidate production solution across multiple cost parameters. That is, a solution is considered a winning production solution if it is no worse than the current solution in all cost parameters and is strictly better than the current solution in at least one cost parameter. The set of inferior production solutions refers to the set of other solutions that are comprehensively superior to the current candidate production solution. That is, the current solution is no worse than these solutions in all cost parameters and is strictly better than them in at least one cost parameter.

[0096] For example, a dominance matrix can be constructed first, and all cost parameters of each pair of options can be compared using a double loop. For each pair of options (A, B), check whether A is no worse than B in all cost parameters and is strictly better than B in at least one parameter. If so, A dominates B, B is added to A's set of inferior production options, and the number of superior production options for B is increased by 1.

[0097] This example quantifies non-dominance relationships into actionable, structured data by explicitly calculating the number of winning and losing production options for each solution. This not only provides the hierarchical ranking process with rigorous quantitative criteria and high transparency but also significantly improves algorithm efficiency.

[0098] S105. Combine the first N production schemes of the multi-level non-dominated set into a new set of parent production schemes. Based on the new set of parent production schemes, return the steps of crossover and mutation operations for iterative processing until the iteration termination condition is met, and obtain the final parent production scheme.

[0099] In this step, the new parent production scheme set refers to the elite population selected from the merged candidate production scheme set for the next iteration. Iteration termination conditions include, but are not limited to, reaching the preset maximum number of iterations, the solution set showing no significant improvement (convergence) in multiple consecutive iterations, or exhaustion of computational resources.

[0100] Combine the first N production schemes of multiple non-dominated sets into a new parent production scheme set, including:

[0101] Starting from the first non-dominated set, the full production plans from each non-dominated set are added to the new parent production plan set in turn;

[0102] During the addition process, determine whether the number of production schemes in the new parent production scheme set has reached N;

[0103] If not, continue adding the full production plans of the next level of non-dominated sets until the number of production plans in the new parent production plan set reaches N.

[0104] In this example, by adding the entire layer of schemes to the new parent set starting from the optimal non-dominated layer (the first layer) in descending order of priority, it is guaranteed that, within the allowable range of numbers, the top-ranked elite individuals will never be replaced by the bottom-ranked individuals, thereby guiding the population to converge quickly and stably toward the Pareto optimal frontier.

[0105] For example, a loop structure can be set up, initially given an empty new parent set and a list of non-dominated layers sorted by priority. The loop starts from the first layer of the list (i.e., the first non-dominated layer). Before adding each layer, it checks if the size of the current new parent set plus the total number of schemes in that layer is less than or equal to N. If it is less than N, all schemes in that layer are added to the new parent set, and the process continues to the next layer. If, when adding a layer (let's say the Kth layer), the total number of schemes exceeds N, the loop terminates. The remaining schemes can be randomly selected from the Kth layer or selected using another algorithm to ensure that the total number of schemes in the new parent set is exactly filled to N.

[0106] This example demonstrates how unconditionally retaining all high-level elite individuals allows for continuous improvement in population performance, resulting in a batch of high-quality production schemes widely distributed across the target space.

[0107] For example, the iteration termination condition includes: the number of iterations reaches a preset number.

[0108] Optionally, the preset number of iterations can be set by conducting a series of small-scale preliminary experiments or by referring to experience with optimization problems of similar scale. For example, observe the curve of solution quality changing with the number of iterations. When the curve enters a plateau and the improvement of the solution set is negligible, the corresponding number of iterations can be used as a reference for the preset number of iterations. For example, the preset number of iterations can be 100.

[0109] The production scheme optimization method provided in this application generates offspring schemes to increase diversity through crossover and mutation operations in a genetic algorithm. Then, using non-dominated sorting, the merged candidate schemes are divided into non-dominated layers of varying quality based on the dominance relationships of multiple cost parameters (such as time and cost). Finally, schemes are selected from high to low priority according to the non-dominated layer, and the distribution of the solution set is ensured within the same layer through crowding calculation. Iterative evolution allows the population to continuously approach the true optimal frontier, ultimately outputting the optimal production scheme. This method can automatically find a batch of optimized schemes that achieve a good balance among multiple competing objectives without artificially weighting multiple objectives into a single objective, providing decision-makers with a rich selection space. Compared with traditional optimization methods, it can ensure that the quality of the solution converges quickly to the Pareto front, and maintain the diversity of the solution set through the crowding mechanism, effectively avoiding getting trapped in local optima. It is particularly suitable for solving multi-objective optimization problems with conflicting objectives in complex production environments.

[0110] This embodiment provides a production scheme optimization method. Figure 2 is a flowchart of the production scheme optimization method provided in this embodiment. As shown in Figure 2, this embodiment, based on the embodiment in Figure 1, provides a detailed description of the process of performing non-dominated hierarchical sorting. The method includes:

[0111] S201. For each candidate production plan in the candidate production plan set, determine the number of winning production plans and the set of losing production plans for each candidate production plan based on the multiple cost parameters of each candidate production plan.

[0112] This step has been described in the foregoing embodiments and will not be repeated here.

[0113] S202. For the first layering operation, the candidate production scheme with zero winning production schemes is determined as the first layer production scheme, and the first layer production scheme is placed into the first layer non-dominated set.

[0114] In this step, the initial stratification operation refers to the first stratification in the non-dominated sorting process, where the top-level non-dominated individuals need to be identified from the complete set of candidate schemes. This step forms the optimal first non-dominated stratum by identifying all individuals not dominated by any other scheme (i.e., candidate production schemes with zero winning production schemes), thus ensuring that the truly elite individuals in the population can be preferentially selected.

[0115] For example, each candidate production plan i in the candidate production plan set has a corresponding number of winning production plans. and a set of disadvantageous production solutions In the initial layering operation, it can be The candidate production solutions are put into the set The remaining candidate production plans are pending further processing.

[0116] S203. For subsequent non-first-time layering operations, based on the set of inferior production schemes corresponding to each upper-level production scheme, the latest number of winning production schemes for each inferior production scheme in the set of inferior production schemes is reduced by one to obtain the updated number of winning production schemes.

[0117] In this step, the non-first-time stratification operation refers to the subsequent stratification iterations performed after the first-level screening is completed. Each iteration processes the remaining unstratified candidate production plans. When a production plan from the previous level is removed, the dominance pressure on the production plans that were originally dominated by them decreases. The subtraction operation accurately reflects this change, providing the correct data basis for identifying the next non-dominated level.

[0118] For example, for Candidate production plan j in the data, i.e. In the context, the set of inferior production solutions for each candidate production solution i. In the candidate production plan j, the number of winning production plans corresponding to each candidate production plan j is determined. .

[0119] S204. Determine the inferior production plan with zero winning production plan as the current layer production plan, and put the current layer production plan into the current layer non-dominated set, until all candidate production plans in the candidate production plan set are put into the corresponding non-dominated set.

[0120] In this step, the entire population is fully hierarchically stratified through iterative processing to ensure that each scheme is assigned to an appropriate non-dominated level, forming a complete Pareto hierarchy structure.

[0121] For example, The candidate production plan j is selected and placed into the set. In the middle, for sets In the context, the set of inferior production plans for each candidate production plan j. Each candidate production scheme H is hierarchically processed in the same way until all candidate production schemes are hierarchically processed and placed into the corresponding non-dominated set.

[0122] The production scheme optimization method provided in this embodiment dynamically updates the changes in the dominance relationship between the remaining schemes by subtracting one from each layer. This avoids the computational overhead of repeated comparisons required in traditional methods, while ensuring the completeness of the layering process and the rigor of the hierarchical structure.

[0123] This embodiment provides a production scheme optimization method. Figure 3 is a flowchart of the production scheme optimization method provided in this embodiment. As shown in Figure 3, this embodiment, based on the embodiment in Figure 2, provides a detailed description of the process of combining the first N production schemes of multiple non-dominated sets into a new parent production scheme set. The method includes:

[0124] S301. Starting from the first layer of non-dominated sets, add the full production plans from each layer of non-dominated sets to the new parent production plan set in sequence.

[0125] S302. If, when adding the Kth layer non-dominated set, all Kth layer production schemes are added to the Kth layer non-dominated set, the number of production schemes in the new parent production scheme set will exceed N, then the Kth layer production schemes in the Kth layer non-dominated set will be sorted by crowding.

[0126] In this step, the Kth level non-dominated set refers to the set of schemes with the Kth priority in the non-dominated ranking, where K is an integer greater than or equal to 1, and the priority decreases as K increases. Crowding ranking is a mechanism that ranks production schemes based on their distribution density in the target space, evaluating their uniqueness by calculating the distance between each production scheme and its neighboring schemes.

[0127] When the number of currently selected solutions is close to but has not yet reached N, a crowding ranking mechanism is introduced to conduct fine-grained screening within the Kth layer. This ensures that the final selected solution not only has a high quality level but also maintains a good distribution in the target space, providing diverse evolutionary material for subsequent iterations.

[0128] For example, we can first calculate the sum of the number of production schemes in the current new parent set and the total number of production schemes in the Kth layer, and compare it with N; when it is predicted that the sum will exceed N, we immediately perform the congestion calculation process, calculate the congestion distance of each production scheme in the Kth layer on each objective function, which is obtained by measuring the sum of the standard deviations of the production scheme and its neighboring schemes in each objective dimension, and finally sort all production schemes in the Kth layer in descending order based on the congestion distance value.

[0129] S303. Based on crowding, select the required number of production schemes for the Kth level from the sorted non-dominated set of the Kth level in order, so that the number of production schemes in the new parent production scheme set is equal to N.

[0130] In this step, by sequentially selecting the required number of individuals from the sorted K-level scheme, a balance is ensured between elite preservation and diversity maintenance in the population, while maintaining a constant population size during algorithm iteration, thus providing a stable environment for evolutionary computation.

[0131] For example, ranking the Kth-level production schemes in the Kth-level non-dominated set by congestion includes:

[0132] Based on the value of each cost parameter corresponding to the K-th level production scheme, an ascending sequence of the K-th level production schemes corresponding to each cost parameter is obtained;

[0133] For the Mth Kth production scheme in a single ascending sequence, calculate the difference in the corresponding cost parameters between the (M-1)th Kth production scheme and the (M+1)th Kth production scheme.

[0134] For each Kth-level production plan, the differences calculated for each cost parameter are summed to obtain the congestion level corresponding to each Kth-level production plan;

[0135] Based on the congestion level of each K-level production scheme, sort the K-level production schemes in the non-dominated set of K-level by congestion level.

[0136] In this example, the uniqueness of each solution is quantified by evaluating the distances of each solution to its nearest neighbor solutions in each objective dimension: first, a sorting sequence is established for each cost parameter; then, the distances between the middle solution and its left and right neighboring solutions are calculated in each sequence; finally, the neighbor distances of each solution in all dimensions are summed to obtain the total crowding degree. The higher the crowding degree of a solution, the sparser its surrounding solution space, and the higher its contribution value to maintaining population diversity.

[0137] For example, for the Mth production plan at the Kth level, the congestion of the current production plan is calculated by summing the distance differences between its two adjacent production plans for each cost parameter. The congestion distance function can be expressed as:

[0138] ;

[0139] in, This represents the value of the (M+1)th production plan at level K for the first cost parameter; This represents the value of the (M-1)th production plan at level K for the first cost parameter; This represents the value of the (M+1)th production plan at level K with respect to the nth cost parameter; This represents the value of the (M-1)th K-th layer production scheme for the nth cost parameter.

[0140] Optionally, the value of the cost parameter can be obtained by selecting the most suitable resource combination for the current production plan, and the minimum value of the cost parameter can be obtained, which is then used for the calculation of congestion.

[0141] For example, suppose a manufacturing company produces multiple products (e.g., mobile phones, computers, tablets, etc.), each requiring multiple manufacturing services (e.g., assembly, testing, packaging, etc.), with each service provided by multiple suppliers (e.g., different assembly plants, testing labs, etc.). Cost parameters could include production time. and production costs It can calculate the minimum production time for the current production plan. And the minimum production cost of the current production plan. These values ​​are then used as their respective cost parameters to calculate congestion. The specific formula is as follows:

[0142]

[0143]

[0144] in, , Indicates the first The product in the first The first manufacturing service Production time of each service provider Indicates the first The product in the first The first service Waiting time for each provider Indicates the first The product in the first The first service Production costs for each provider.

[0145] This example, by determining the local distribution density of each production scheme across all objective dimensions, can accurately identify unique individuals located in sparse regions. These individuals are then prioritized for retention during the selection process, effectively preventing the over-concentration of excellent but similar production schemes. This ensures that the final set of selected production schemes is not only of superior quality but also broadly covers different trade-offs across all objective dimensions, thus providing a more representative Pareto optimal solution set.

[0146] The production scheme optimization method provided in this embodiment is based on the convergence advantage of the elite retention strategy and avoids excessive aggregation of the population in local areas by sorting by crowding degree. As a result, it can obtain a Pareto optimal solution set with uniform distribution and wide coverage in complex multi-objective optimization problems, which significantly improves the quality and practicality of the optimization results.

[0147] Figure 4 is a production and manufacturing process flowchart provided in an embodiment of the present invention. Taking the manufacturing process of customized clothing as an example: First, it can be achieved through intelligent measurement services. And personalized demand collection services Collect data, and then process it through the order management service. Orders are generated. Next, the automated design service phase begins, which includes: apparel design services. Songliang Service Drawing services and intelligent material scheduling services In manufacturing services This includes tailoring services. Sewing services Ironing service Quality inspection services and packaging services Finally, through logistics services Send the customized clothes to the customer.

[0148] Assume the current order requires the production of products A, B, and C, and ignore intelligent body measurement services. Personalized demand collection service and logistics services Production time and costs. Order management services for various products. Clothing design services , loose volume service CAD drawing services Garment layout service Manufacturing services The specific parameter settings are shown in Table 1.

[0149]

[0150] First, arrange each service resource in sequence. , , , , ...sort them, where Let $j$ represent the $i$-th service and $j$-th supplier. Each service is coded, with a value of 1 if selected and 0 otherwise. Then, based on the order quantity $S$, $S$ service sequences are randomly generated; each service sequence represents a production plan. The crossover and mutation probabilities can be set to 100%. S more service sequences are generated through crossover and mutation. Furthermore, a fast non-dominated sorting algorithm is used to retain the production service sequence with the optimal production time and cost in each iteration, i.e., the production plan. The number of iterations in the experiment can be set to 100.

[0151] Assume the production quantities of products A and B are fixed at 50, while the quantity of product C varies depending on the quantity of products in the order. Figure 5 shows the distribution of the shortest production time for the reorganized production schemes obtained by three different calculation strategies; Figure 6 shows the distribution of the lowest production cost for the reorganized production schemes obtained by three different calculation strategies.

[0152] Experimental results show that, among the three resource allocation schemes—the production scheme optimization method, genetic algorithm, and random strategy—provided in this application, the production scheme optimization method can achieve the resource allocation strategy with the lowest production time and production cost.

[0153] This embodiment also provides a production scheme optimization device. Figure 7 is a diagram of a production scheme optimization device provided in this embodiment of the invention. As shown in Figure 7, the production scheme optimization device 70 includes:

[0154] The acquisition module 701 is used to acquire the parent production plan set, which includes multiple production plans for the same production task, and each production plan includes multiple cost parameters.

[0155] Execution module 702 is used to perform crossover and mutation operations on the parent production scheme set to obtain the child production scheme set;

[0156] The processing module 703 is used to merge the parent production scheme set and the child production scheme set to obtain a candidate production scheme set, and to perform non-dominated hierarchical sorting of the candidate production schemes based on multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set to obtain a multi-level non-dominated set. The priority of the non-dominated set decreases sequentially as the number of levels increases.

[0157] The iteration module 704 is used to combine the first N production schemes of the multi-level non-dominated set into a new set of parent production schemes. Based on the new set of parent production schemes, the steps of crossover and mutation operations are returned for iterative processing until the iteration termination condition is met, and the final parent production scheme is obtained.

[0158] In one possible implementation, the processing module 703 is specifically used for:

[0159] For each candidate production plan in the candidate production plan set, based on multiple cost parameters of each candidate production plan, determine the number of winning production plans and the set of inferior production plans for each candidate production plan. The number of winning production plans for each candidate production plan is the number of winning production plans whose cost parameters are all better than the corresponding candidate production plan. The cost parameters of each candidate production plan are better than each inferior production plan in the set of inferior production plans.

[0160] Based on the number of winning production schemes corresponding to each candidate production scheme and the set of losing production schemes, the candidate production schemes in the candidate production scheme set are sorted in a non-dominated hierarchical manner to obtain a multi-level non-dominated set.

[0161] In one possible implementation, the processing module 703 is specifically used for:

[0162] For the first stratification operation, the candidate production schemes with zero winning production schemes are identified as the first-level production schemes, and the first-level production schemes are placed into the first-level non-dominated set.

[0163] For subsequent non-first-time stratification operations, based on the set of inferior production schemes corresponding to each production scheme in the upper layer, the latest number of superior production schemes for each inferior production scheme in the set of inferior production schemes is reduced by one to obtain the updated number of superior production schemes.

[0164] The inferior production plan with zero winning production plan after the update is identified as the current layer production plan, and the current layer production plan is placed into the current layer non-dominated set, until all candidate production plans in the candidate production plan set are placed into the corresponding non-dominated set.

[0165] In one possible implementation, the iteration module 704 is specifically used for:

[0166] Starting from the first non-dominated set, the full production plans from each non-dominated set are added to the new parent production plan set in turn;

[0167] During the addition process, determine whether the number of production schemes in the new parent production scheme set has reached N;

[0168] If not, continue adding the full production plans of the next level of non-dominated sets until the number of production plans in the new parent production plan set reaches N.

[0169] In one possible implementation, the iteration module 704 is also used for:

[0170] If, when adding the Kth level non-dominated set, all Kth level production schemes of the Kth level non-dominated set are added, the number of production schemes in the new parent production scheme set will exceed N, then the Kth level production schemes in the Kth level non-dominated set will be sorted by crowding.

[0171] Based on crowding order, select the required number of production schemes for the Kth level from the sorted non-dominated set of the Kth level in order, so that the number of production schemes in the new parent production scheme set is equal to N.

[0172] In one possible implementation, the iteration module 704 is specifically used for:

[0173] Based on the value of each cost parameter corresponding to the K-th level production scheme, an ascending sequence of the K-th level production schemes corresponding to each cost parameter is obtained;

[0174] For the Mth Kth production scheme in a single ascending sequence, calculate the difference in corresponding cost parameters between the (M-1)th Kth production scheme and the (M+1)th Kth production scheme;

[0175] For each Kth-level production plan, the differences calculated for each cost parameter are summed to obtain the congestion level corresponding to each Kth-level production plan;

[0176] Based on the congestion level of each K-level production scheme, sort the K-level production schemes in the non-dominated set of K-level by congestion level.

[0177] In one possible implementation, the iteration module 704 includes an iteration termination condition where the number of iterations reaches a preset number.

[0178] This embodiment provides a production scheme optimization device that can execute the production scheme optimization method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0179] Figure 8 is a hardware schematic diagram of an electronic device provided in an embodiment of the present invention. As shown in Figure 8, the electronic device 80 provided in this embodiment includes at least one processor 801 and a memory 802. The device 80 also includes a communication component 803. The processor 801, the memory 802, and the communication component 803 are connected via a bus 804.

[0180] In the specific implementation process, at least one processor 801 executes computer execution instructions stored in memory 802, causing at least one processor 801 to perform the above method.

[0181] The specific implementation process of processor 801 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0182] In the embodiment shown in Figure 8 above, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0183] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0184] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0185] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the method described above.

[0186] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0187] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0188] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0189] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0190] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0191] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0192] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0193] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application 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 or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for optimizing a production scheme, characterized in that, The method includes: obtaining a parent production scheme set, the parent production scheme set including multiple production schemes for the same production task, each production scheme including multiple cost parameters; performing crossover and mutation operations on the parent production scheme set to obtain a child production scheme set; merging the parent production scheme set and the child production scheme set to obtain a candidate production scheme set, and performing non-dominated hierarchical sorting on the candidate production schemes based on the multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set to obtain a multi-level non-dominated set, the priority of the non-dominated set decreasing sequentially with the number of levels; combining the first N production schemes of the multi-level non-dominated set into a new parent production scheme set; and based on the new parent production scheme set, returning to the crossover and mutation operation steps for iterative processing until the iteration termination condition is met to obtain the final parent production scheme.

2. The method according to claim 1, characterized in that, The step of performing non-dominated hierarchical sorting of candidate production schemes based on multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set includes: for each candidate production scheme in the candidate production scheme set, determining the number of winning production schemes and the set of losing production schemes corresponding to each candidate production scheme according to multiple cost parameters of each candidate production scheme, wherein the number of winning production schemes corresponding to each candidate production scheme is the number of winning production schemes whose cost parameters are all better than the corresponding candidate production scheme, and the cost parameters of each candidate production scheme are all better than each losing production scheme in the corresponding set of losing production schemes; and performing non-dominated hierarchical sorting of candidate production schemes in the candidate production scheme set according to the number of winning production schemes and the set of losing production schemes corresponding to each candidate production scheme to obtain a multi-level non-dominated set.

3. The method according to claim 2, characterized in that, The step of performing non-dominated hierarchical sorting of candidate production schemes in the candidate production scheme set based on the number of winning production schemes corresponding to each candidate production scheme and the set of inferior production schemes to obtain multi-level non-dominated sets includes: for the first hierarchical operation, determining the candidate production schemes with zero winning production schemes as the first-level production schemes and placing the first-level production schemes into the first-level non-dominated set; for subsequent non-first-level hierarchical operations, based on the set of inferior production schemes corresponding to each production scheme in the previous level, decrementing the latest number of winning production schemes for each inferior production scheme in the inferior production scheme set by one to obtain an updated number of winning production schemes; determining the inferior production schemes with zero updated number of winning production schemes as the current-level production schemes and placing the current-level production schemes into the current-level non-dominated set, until all candidate production schemes in the candidate production scheme set are placed into their corresponding non-dominated sets.

4. The method according to claim 1, characterized in that, The step of combining the first N production schemes of multiple non-dominated sets into a new parent production scheme set includes: starting from the first layer of non-dominated sets, sequentially adding all production schemes from each layer of non-dominated sets to the new parent production scheme set; during the addition process, determining whether the number of production schemes in the new parent production scheme set has reached N; if not, continuing to add all production schemes from the next layer of non-dominated sets until the number of production schemes in the new parent production scheme set reaches N.

5. The method according to claim 4, characterized in that, The method further includes: if, when adding the Kth layer non-dominated set, adding all the Kth layer production schemes of the Kth layer non-dominated set would result in the number of production schemes in the new parent production scheme set exceeding N, then the Kth layer production schemes in the Kth layer non-dominated set are sorted by congestion; based on the congestion sort, the required number of Kth layer production schemes are selected sequentially from the sorted Kth layer non-dominated set so that the number of production schemes in the new parent production scheme set is equal to N.

6. The method according to claim 5, characterized in that, The step of sorting the Kth-level production schemes in the Kth-level non-dominated set by congestion includes: obtaining an ascending sequence of Kth-level production schemes for each cost parameter based on the value of each cost parameter corresponding to the Kth-level production scheme; for the Mth Kth-level production scheme in a single ascending sequence, calculating the difference between the (M-1)th and (M+1)th Kth-level production schemes on the corresponding cost parameter; for each Kth-level production scheme, summing the differences calculated on each cost parameter to obtain the congestion degree corresponding to each Kth-level production scheme; and sorting the Kth-level production schemes in the Kth-level non-dominated set by congestion degree based on the congestion degree corresponding to each Kth-level production scheme.

7. The method according to claim 1, characterized in that, The iteration termination condition includes: the number of iterations reaches a preset number.

8. A production scheme optimization device, characterized in that, The apparatus includes: an acquisition module for acquiring a set of parent production schemes, the set of parent production schemes including multiple production schemes for the same production task, each production scheme including multiple cost parameters; an execution module for performing crossover and mutation operations on the set of parent production schemes to obtain a set of child production schemes; a processing module for merging the set of parent production schemes and the set of child production schemes to obtain a set of candidate production schemes, and performing non-dominated hierarchical sorting on the candidate production schemes based on the multiple cost parameters corresponding to each candidate production scheme in the candidate production scheme set to obtain a multi-level non-dominated set, the priority of the non-dominated set decreasing sequentially with the number of levels; and an iteration module for combining the first N production schemes of the multi-level non-dominated sets into a new set of parent production schemes, and based on the new set of parent production schemes, returning to the steps of the crossover and mutation operations for iterative processing until the iteration termination condition is met to obtain the final parent production scheme.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: At least one processor and a memory; wherein the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, causing the at least one processor to perform the method as claimed in any one of claims 1-7.

Images