Productivity optimization device and method for multi-level production scheduling

Abstract

The invention provides a productivity optimization device and method for a multi-level production schedule, and relates to the technical field of production line data processing, and the method comprises a combination constraint setting module which is used for setting a plurality of combination constraint factors for the multi-level production schedule; the node set determination module is used for outputting multiple groups of node sets; the bearing performance analysis module is used for obtaining a bearing performance analysis result according to the bearing performance analysis model; the node set combination module is used for acquiring a multi-combination union set; and the scheduling balance scheduling module is used for outputting a global balance strategy. According to the method and the device, the technical problems of resource waste and low data processing efficiency caused by independent calculation of production tasks in complex production lines can be solved, and the technical effects of reducing the data processing amount, improving the resource utilization rate and reducing the productivity waste by analyzing the carrying performance of each production line and pre-combining the production lines into a whole are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of production line data processing, and in particular to a capacity optimization device and method for multi-level production scheduling. Background Art

[0002] With the development of manufacturing industry, production scheduling and capacity management have become particularly important in modern enterprises. Especially in the production environment of multiple varieties and small batches, reasonable production scheduling not only directly affects the production efficiency of enterprises, but also relates to the timeliness of order delivery and the rational use of production resources. However, traditional production scheduling methods usually rely on fixed scheduling modes and limited resource allocation methods, which are difficult to effectively deal with the uncertainty and complexity in the production environment. At present, enterprises face many challenges in the production process, such as: diversified production tasks, shortened order delivery time, uneven capacity demand, low equipment utilization, etc. In order to meet these challenges, many enterprises have introduced production scheduling systems to improve resource utilization. However, the existing production scheduling systems generally have the problems of single scheduling level and lack of dynamic adjustment capabilities, which makes it difficult to adapt to real-time changes in complex production environments. This makes the optimization of enterprise capacity a major problem, often resulting in waste of resources, idle equipment or low production efficiency. Summary of the invention

[0003] The purpose of this application is to provide a capacity optimization device and method for multi-level production scheduling, so as to solve the technical problems of resource waste and low data processing efficiency caused by independent calculation of production tasks in complex production lines.

[0004] In view of the above problems, the present application provides a capacity optimization device and method for multi-level production scheduling.

[0005] In the first aspect, the present application also provides a capacity optimization device for multi-level production scheduling, the device comprising: a combination constraint setting module, used to obtain a multi-level production schedule, and set multiple combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the number of combinations of the corresponding level production schedule; a node set determination module, used to determine the node set of each level production schedule in the multi-level production schedule, and output multiple groups of node sets; a continuity analysis module, used to input the information of each group of node sets into a continuity analysis model, and obtain a continuity analysis result according to the continuity analysis model; a node set combination module, used to combine each group of node sets with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple groups of merged sets, wherein each level of production schedule corresponds to a group of merged sets; a scheduling balance scheduling module, used to determine the capacity target, perform scheduling balance scheduling on the multiple groups of merged sets with the capacity target, and output a global balance strategy.

[0006] In a second aspect, the present application provides a method for optimizing the capacity of a multi-level production schedule, and the method is implemented by a device for optimizing the capacity of a multi-level production schedule described in the first method, wherein the method includes: obtaining a multi-level production schedule, setting multiple combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the number of combinations of the corresponding level production schedule; determining a node set of each level of the production schedule in the multi-level production schedule, and outputting multiple groups of node sets; inputting information of each group of node sets into a continuity analysis model, and obtaining a continuity analysis result according to the continuity analysis model; combining each group of node sets with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple groups of merged sets, wherein each level of the production schedule corresponds to a group of merged sets; determining a capacity target, scheduling and balancing the multiple groups of merged sets with the capacity target, and outputting a global balancing strategy.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The device provided in the embodiment of the present application obtains a multi-level production schedule through a combined constraint setting module, sets multiple combined constraint factors for the multi-level production schedule, wherein each combined constraint factor is used to constrain the number of combinations of the corresponding level production schedule; determines the node set of each level production schedule in the multi-level production schedule through a node set determination module, and outputs multiple groups of node sets; inputs the information of each group of node sets into a continuity analysis model through a continuity analysis module, and obtains a continuity analysis result according to the continuity analysis model; and uses the node set combination module to combine the multiple combined constraint factors as constraints. Conditions, each group of node sets is combined with the continuity analysis result as input variables to obtain multiple combined sets, wherein each level of production schedule corresponds to a combined set; through the scheduling balance scheduling module, the capacity target is determined, the multiple combined sets are scheduled in a balanced manner with the capacity target, and a global balancing strategy is output, which effectively solves the technical problems of resource waste and low data processing efficiency caused by independent calculation of production tasks in complex production lines, and achieves the technical effect of reducing data processing volume, improving resource utilization and reducing capacity waste by performing continuity analysis on each production line and pre-combining the production lines as a whole.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a schematic diagram of the structure of a capacity optimization device for multi-level production scheduling in this application; Figure 2 A flow chart of a method for optimizing the production capacity of a multi-level production scheduling system is provided in this application.

[0011] Explanation of the reference numerals: combined constraint setting module 1, node set determination module 2, continuity analysis module 3, node set combination module 4, schedule balance scheduling module 5. DETAILED DESCRIPTION

[0012] The present application solves the technical problems of resource waste and low data processing efficiency caused by independent calculation of production tasks in complex production lines by providing a capacity optimization device and method for multi-level production scheduling, and achieves the technical effect of reducing data processing volume, improving resource utilization and reducing capacity waste by conducting continuity analysis on each production line and pre-combining the production lines as a whole.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] For example, please refer to the attached Figure 1 The present application provides a capacity optimization device for multi-level production scheduling, and the device specifically comprises the following steps: The combination constraint setting module 1 is used to obtain a multi-level production schedule and set a plurality of combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the combination quantity of the corresponding level production schedule.

[0015] Specifically, in the combination constraint setting module 1, in the multi-level production scheduling, in order to ensure that the scheduling is more accurate and effective, multiple combination constraint factors are set for the production tasks at each level. These combination constraint factors limit the number of combinations at each level when scheduling, so that the production arrangement at each level not only meets the current demand, but also optimizes the utilization of resources. The combination constraint factor is a restriction condition formulated according to the number of collaborative nodes at each level. For example, at the workshop level, a combination constraint factor can be set to limit the number of production tasks of each device within a specific time; at the process level, a constraint factor can be set to limit the number of parallel tasks of a specific process. These factors ensure that resources are not overloaded or idle, while ensuring the feasibility of production scheduling. Through such a multi-level constraint design, the number of schedules at each production level is controlled, avoiding too many or too few combinations, which helps to achieve the optimal configuration of the overall production capacity.

[0016] Furthermore, multiple combined constraint factors are set for the multi-level production schedule, including: Collect historical node collaboration data sets for each level of production scheduling; extract the number of collaboration nodes from the historical node collaboration data sets for each level of production scheduling, and output a sample of the number of collaboration nodes; obtain the preferred number of collaboration nodes for each level of production scheduling based on the sample of the number of collaboration nodes, wherein the preferred number of collaboration nodes is the number with the highest frequency; set multiple combination constraint factors based on the preferred number of collaboration nodes for each level of production scheduling.

[0017] In a preferred embodiment, at each level of production scheduling (e.g., workshop, process, equipment level), a collaborative data set of each historical node is collected. The historical node refers to a specific step or event that occurs in each production scheduling process, and these nodes contain data on how each level collaborates in actual production. These data may include the usage of each device at a specific time, the task allocation of different processes, and the collaborative relationship between resources (such as personnel and machines). Subsequently, the number of nodes is extracted from the production scheduling data set of each level, and the number of collaborations of each historical node in each level, that is, the actual number of collaborations between different nodes, is calculated, and these numbers are output as collaborative node number samples. These samples record the scale of collaboration at different production levels when completing tasks in the actual production process. Afterwards, based on the collaborative node number samples of each level, the most frequent collaborative number is analyzed and found, and it is used as the preferred collaborative node number. This preferred collaborative node number represents the most common collaborative scale in the production level, reflecting the number of collaborative nodes required for the production scheduling of this level in most cases. Then, according to the preferred collaborative node number of each level of production scheduling, multiple combined constraint factors are set for the production scheduling. These constraint factors will limit the number of node combinations for each level of production scheduling, using the preferred number of collaborations as the standard to ensure that the production arrangements at each level meet the most common and optimal number of collaboration nodes. This setting helps optimize scheduling, enabling each level to effectively collaborate when performing production tasks and improve overall production efficiency. Through this process, the scheduling of each production level obtains the optimal number of node combinations based on historical collaboration data, providing a more reasonable constraint setting for the scheduling system, thereby allocating resources and arranging production tasks more efficiently.

[0018] The node set determination module 2 is used to determine the node set of each level of production schedule in the multi-level production schedule and output multiple groups of node sets.

[0019] Specifically, in the node set determination module 2, in order to effectively organize and manage the production tasks at each level, first, the production scheduling requirements at each level are identified from the overall production schedule. Each level may correspond to a different range of production tasks, such as equipment level, process level, workshop level, etc. Identifying these levels helps to clarify the node types required at each level. According to different level requirements, the production node information involved in each level is collected. The node information includes key task points, resource points and process nodes in the production schedule of this level. For example, the nodes at the equipment level can be specific production equipment; the nodes at the process level can be operations such as cutting, assembly, and testing; and the nodes at the workshop level can be the process sets of the production line or region. The production node information collected at each level is grouped by level to construct the corresponding node set. This step ensures that the production nodes at each level are organized and divided into corresponding sets. In this way, after the nodes at each production level are divided according to structure and functionality, multiple sets of node sets can be output, each set representing a production task at a different level. In this way, the production schedules at each level can operate independently, and at the same time, the overall production optimization and reasonable resource allocation can be achieved through the mutual cooperation between the node sets. These multiple sets of node collections provide a data basis for further scheduling optimization and dynamic adjustment, helping to improve the response speed and efficiency of the entire production system.

[0020] The continuity analysis module 3 is used to input the information of each group of node sets into the continuity analysis model, and obtain the continuity analysis result according to the continuity analysis model.

[0021] Specifically, in the continuity analysis module 3, after each group of node sets is determined, the information of these node sets is input into the pre-built continuity analysis model to evaluate the continuity and coordination between each production node. The continuity analysis model is mainly used to analyze the connection of each group of nodes in the production process to ensure that the combined nodes will not cause excessive switching time and resource waste when connected. The detailed information of each group of node sets is input into the continuity analysis model. This information includes key parameters such as the task type, resource requirements, and time arrangement of the node. Through the model, the continuity of these nodes in the production process can be analyzed, and the connection between the nodes can be calculated one by one according to the learned mapping relationship, and the time interval between the nodes, the matching degree of resource requirements, and the task dependency can be checked to evaluate the continuity characteristics between the nodes. For example, the model will evaluate whether the conditions are met to smoothly transfer the task to the next process node after a certain production line node is completed. Subsequently, the continuity characteristics between the nodes generated by the continuity analysis model are weighted and calculated to generate the continuity analysis results, which provide a reference for scheduling optimization, so that the node configuration and resource arrangement can be further adjusted. Through this process, the continuity analysis model can help identify and solve connection problems in production scheduling, ensure efficient connection and resource optimization of each node in the entire production process, and ultimately improve the overall efficiency and stability of the production system.

[0022] Further, obtaining the continuity analysis result according to the continuity analysis model includes: The continuity analysis model analyzes the information of each group of node sets, calculates the continuity characteristics between different nodes, and the continuity characteristics between nodes include the continuity between nodes, the switching efficiency between nodes, and the resource compatibility between nodes; constructs a feature matrix based on the continuity characteristics between different nodes; establishes a weight matrix corresponding to the feature matrix, and performs matrix calculation on the feature matrix based on the weight matrix to obtain the continuity analysis result.

[0023] In a preferred embodiment, the information of each set of node sets is input into the continuity analysis model, and the model calculates the continuity characteristics between different nodes according to the learned mapping relationship. The continuity characteristics include the continuity between nodes, the switching efficiency between nodes, and the resource compatibility between nodes. Among them, the continuity between nodes is determined by checking whether the output content of the previous node can be used as the input of the next node through the continuous analysis channel inside the model, ensuring the smooth connection of the process. For example, the output material or product of the previous production line should be the input content of the next production line, so as to ensure the compatibility between nodes. This feature ensures the continuity and seamless connection of the production process. The switching efficiency between nodes is determined by evaluating the overhead of tasks when switching between different nodes through the frequency analysis channel inside the model. Higher switching efficiency means smoother conversion between nodes, reducing waiting or stagnation caused by poor connection. The resource compatibility between nodes is determined by analyzing the matching degree of each node in resource allocation through the compatible analysis channel inside the model, ensuring that the resources (such as equipment, manpower, and materials) between nodes are compatible. For example, can the semi-finished product completed by the previous node be directly processed or used by the next node, thereby avoiding pauses or repeated configurations caused by resource incompatibility? According to the continuity, switching efficiency and resource compatibility characteristics between nodes calculated by the model, these characteristics are organized into a feature matrix according to the relationship between nodes. Each element in the matrix represents a certain continuity feature between two specific nodes, which is convenient for subsequent comprehensive analysis. For example, a row of the feature matrix can represent the connection between a certain node and multiple subsequent nodes, and each column represents the specific continuity feature value. Subsequently, according to the priority of the production system for different continuity features, a one-to-one corresponding weight matrix is ​​constructed based on the feature matrix. Each element value of the weight matrix reflects the importance of the continuity feature at the corresponding position in the feature matrix in the overall continuity analysis, ensuring that the analysis results meet actual needs. Afterwards, a comprehensive analysis result matrix is ​​obtained by performing matrix operations (such as matrix multiplication) on the feature matrix and the weight matrix, and this analysis result matrix is ​​output as the continuity analysis result. The result reflects the level of connectivity between each pair of nodes, including the consistency of connection, resource utilization efficiency and the smoothness of task switching, providing a basis for further optimization of scheduling, enabling precise adjustments to the connectivity characteristics between nodes, ensuring that the entire production process remains efficient and stable in terms of node connections.

[0024] For the continuity analysis model, the continuous analysis channel, frequency analysis channel, and compatible analysis channel are constructed respectively, and these three channels are integrated to form this continuity analysis model. For the continuous analysis channel, the construction methods include support vector machine (SVM), random forest, etc. Taking SVM as an example, first collect the historical node set information, extract the historical material specification matching, historical time interval and other features from the historical information of each pair of nodes, and combine these features into feature vectors, and then continuously identify each feature vector. At the same time, define the target variable, and use the binary classification label to represent the continuity state of the node pair, where 1 represents high continuity and 0 represents low continuity. When the SVM model is initialized, the RBF kernel (Gaussian kernel) is selected because the kernel function can map nonlinear data to a higher dimension, making it easier to find a suitable classification boundary. In order to ensure the accuracy and stability of the model, the hyperparameters of the model are optimized using cross-validation before the formal training of the model. The penalty coefficient C determines the tolerance of the model to misclassification, while the γ parameter controls the influence range of each sample on the decision boundary. Through multiple cross-validations, the optimal combination of C and γ values ​​is found to achieve the best classification effect. During the formal training process, the transformed feature vectors are input into the model, and the SVM finds the optimal classification boundary by minimizing the loss function. The model identifies the key data points closest to the boundary, which are called support vectors, and they directly affect the location of the classification boundary. By introducing the Lagrange multiplier method, the SVM converts the complex optimization problem into a dual problem, which can more efficiently calculate the best hyperplane that meets all classification requirements. In each iteration, the model calculates the gradient of the loss function and gradually adjusts the weights and biases using gradient descent or other optimization algorithms so that the loss function gradually converges to the minimum value. The support vectors are also dynamically updated in each iteration to ensure that the classification boundary is continuously optimized. After multiple iterations, the model gradually approaches the optimal separation boundary by gradually adjusting the boundary and support vectors. During the training process, the model constantly checks which data points should be used as support vectors, thereby forming a stable set of support vectors and decision boundaries. When the support vectors and decision boundaries of the model no longer change significantly, it means that the model has converged. At this time, the SVM has completed training and obtained a continuous analysis channel that can effectively classify high-continuity and low-continuity node pairs. For the frequency analysis channel and the compatible analysis channel, a neural network is used to construct them in a similar way to the above. Through forward propagation, loss calculation, back propagation, parameter optimization and other steps, the frequency analysis channel and the compatible analysis channel are gradually trained so that they approach convergence.

[0025] The node set combination module 4 is used to combine each group of node sets with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple groups of combined sets, wherein each level of production scheduling corresponds to a group of combined sets; Specifically, in the node set combination module 4, multiple combination constraint factors are used as constraint conditions, combined with the continuity analysis results, to optimize the combination of each group of node sets. The combination constraint factor is used to limit the number of nodes allowed in each combination. The continuity analysis results provide the connection between each pair of nodes, and this information helps to judge the rationality and feasibility of the node combination. Based on these constraints and analysis results, a corresponding combination will be generated for each level of production scheduling. For each level of node set, by analyzing the continuity characteristics, the nodes that meet the constraints are combined together to form multiple possible combinations. Each combination represents a reasonable collaboration mode of different production nodes at this level, ensuring that the resource usage and connection relationship of each node meet the established requirements. Finally, multiple groups of merged sets are obtained, in which each level of production scheduling corresponds to a group of merged sets, providing a clear framework for subsequent scheduling optimization.

[0026] Furthermore, each group of node sets is combined with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple combined sets, including: Obtain node pairs with a connectivity greater than a preset connectivity in the connectivity analysis results; combine the node pairs with a connectivity greater than the preset connectivity to obtain an initial merge group; record the current number of nodes in each initial merge group, and if the current number of nodes in each initial merge group is greater than or equal to the corresponding value in the multiple combination constraint factors, obtain a multiple-group merge set.

[0027] In a preferred embodiment, the connectability analysis result includes the connection score of each node pair, that is, the comprehensive score of continuity, switching efficiency and resource compatibility between node pairs obtained by weighted calculation of the continuity, switching efficiency and resource compatibility of the node pair. These scores are compared with the preset connectability threshold, and node pairs with scores higher than the threshold are selected, that is, those node pairs showing good connectability. After screening out the node pairs that meet the requirements, the node pairs of each layer are combined to form an initial merge group. The initial merge group is composed of these high-connectivity node pairs, and each group contains a node combination with smooth connection and resource compatibility. These initial merge groups are the basis for subsequent combined merge sets, ensuring that the nodes in each group can work together efficiently in actual production. Subsequently, the number of nodes currently contained in each initial merge group is recorded, and the number of nodes is compared with the corresponding constraints in the combined constraint factor to ensure that the scale of each initial merge group meets the preset constraints. The combined constraint factor defines the lower limit of the number of nodes allowed in each group to prevent too few nodes in a single group, thereby affecting the resource allocation and overall efficiency of production. If the number of nodes in the initial merged group is greater than or equal to the number specified in the combination constraint factor, the group meets the conditions and is marked as a valid combination. When all initial merged groups have been verified for the number of nodes and meet the constraint factor requirements, these combinations with valid combination marks are integrated into a multi-group merge set. The final multi-group merge set contains the valid node combinations in each level of production scheduling, ensuring that each group of nodes has both high connectivity and meets the scale constraints, providing a reasonable node combination solution for the further execution of the production plan.

[0028] Furthermore, if the current number of nodes in each initial merge group is less than the corresponding value in the multiple combination constraint factors, a secondary combination instruction is obtained; according to the secondary combination instruction, the continuity analysis result corresponding to the initial merge group is recalculated; and the initial merge group is secondary combined according to the recalculated continuity analysis result until the current number of nodes in each initial merge group is greater than or equal to the corresponding value in the multiple combination constraint factors.

[0029] In a feasible implementation, after each initial merge group is generated, the current number of nodes in each group is checked. If the number of nodes in a certain initial merge group is less than the number of nodes set in the combined constraint factor, it is determined that the group does not meet the constraint requirements, and at this time, a secondary combination instruction is triggered. According to the secondary combination instruction, the continuity analysis results between the nodes in the initial merge group are recalculated in the same manner as described above to further find node pairs that meet the combination conditions. In this process, the continuity of the node pairs in the initial merge group is re-evaluated, including recalculating the continuity between the nodes, switching efficiency, and resource compatibility. After obtaining the recalculated continuity analysis results, the initial merge group is secondary combined, and the node pairs that meet the preset continuity are gradually added to the current merge group, and the number of nodes after the combination is continuously checked to meet the constraint factor requirements. Repeat the above process until the number of nodes in each initial merge group reaches or exceeds the number specified in the combined constraint factor. Once the number of nodes in all initial merge groups meets the requirements, these adjusted merge groups will be regarded as valid combinations, and finally a multi-group merge set that meets all constraints is formed.

[0030] The scheduling and balancing dispatching module 5 is used to determine the capacity target, perform scheduling and balancing dispatching on the plurality of group combination sets according to the capacity target, and output a global balancing strategy.

[0031] Specifically, in the scheduling balanced scheduling module 5, in the production scheduling process, the specific capacity target is first determined, that is, the production output that is expected to be achieved within a given time. The determination of the capacity target provides a clear direction for subsequent scheduling and scheduling, so that the scheduling can be optimized around this target. After clarifying the capacity target, the production tasks at each level are scheduled and balanced according to the node combination of each group in the multi-group merge set, and the capacity load of each merge group is evaluated to ensure that the production resources are reasonably allocated among the tasks. The key to balanced scheduling is to avoid overload or resource waste of individual nodes by optimizing resource allocation and time scheduling, thereby improving the overall production efficiency and capacity utilization. After completing the scheduling balance of all node groups, a global equilibrium strategy is output. This global strategy not only meets the capacity target, but also ensures that the resource allocation of each production link is reasonable and the task arrangement is smooth, so as to achieve the smooth operation of the overall production process. The final global equilibrium strategy will serve as the core guidance plan for production execution to ensure the realization of capacity targets and efficient use of resources.

[0032] Furthermore, the multi-group combined set is scheduled and balanced according to the production capacity target, and a global balancing strategy is output, including: Receive the capacity target, which includes a preset load variance; perform capacity load evaluation on each merged group in each merged set, and output the load evaluation data of each merged group; construct a balancing objective function based on the capacity target, call the balancing objective function to perform scheduling balancing response to the load evaluation data of each merged group, and output a global balancing strategy, wherein the global balancing strategy includes multiple balancing strategies corresponding to the multi-level production schedule.

[0033] In a preferred embodiment, when performing global balanced scheduling, the capacity target is first received. The capacity target not only includes the production volume requirement, but also includes a preset load variance. This load variance is used to measure the degree of load balance between different merged groups. The goal is to make the load distribution of each merged group as close as possible while meeting the capacity, so as to avoid overload or waste of resources in some groups. After obtaining the preset load variance, the capacity load of each merged group in the merged set is evaluated. The load evaluation includes calculating the current capacity load of each merged group, that is, the resource usage of each group under the current task allocation, which is obtained by calculating the ratio of the resources consumed by the current task to the available resources, thereby forming a set of load evaluation data. Subsequently, a balanced objective function is constructed based on the capacity target and the preset load variance. The core of this balanced objective function is to minimize the load difference between the merged groups, so that the load of each group is as close as possible to the range of the preset load variance. The balanced objective function is as follows: ; Where f is the value of the objective function, which represents the sum of the load variance and the penalty term. The optimization goal is to minimize the value of f to achieve balanced load distribution. is the load assessment data for the ith merged group. N is the total number of merged groups, which is used to calculate the mean load and load variance. is the average load of the merged group, which represents the average evaluation data of all merged groups and is used to evaluate the difference between each merged group and the average load. T is the preset load variance, which represents the tolerance for load imbalance. A lower T value requires a more balanced load, while a higher T value allows for greater fluctuations in the load between merged groups. is the penalty coefficient, which is used to control the impact of the penalty term when the load variance exceeds T. The value will tend to meet the upper limit of the variance requirement, while smaller The value will weaken the impact of the penalty term. is a penalty term. It will take effect only when the load variance exceeds the set variance upper limit T. It is used to constrain the load variance and avoid exceeding the target value T. The balance objective function will comprehensively consider the load evaluation data of each merged group to ensure that the load is evenly distributed under the premise of meeting the capacity requirements. After that, the balance objective function is called to schedule the load evaluation data of each merged group. According to the balance objective function, the task allocation or resource allocation of each merged group is adjusted to make the load more evenly distributed among different merged groups. This process includes adjusting the production priority of each merged group, reallocating some resources, or changing the task scheduling order to ensure that the load difference meets the preset variance requirements. After completing the scheduling balance response, the global balance strategy is generated and output. The global balance strategy includes the balance plan of production scheduling at each level to ensure that all production links at multiple levels have reached a balanced load state. Each balance strategy describes in detail the tasks, resource allocation and scheduling order of each merged group to ensure the realization of the capacity target and the reasonable distribution of the load. Through this series of steps, the response to the capacity target is achieved, and the global balanced production scheduling strategy is achieved, so that the entire production system can operate efficiently and evenly.

[0034] Furthermore, multiple groups of merged sets are obtained, including: The average node load of each merged group in each merged set is obtained; according to the average node load of each merged group, abnormal merged groups with load deviation are identified, the abnormal merged groups are reorganized, and the multiple merged sets are updated.

[0035] In an optional implementation, the load information of each node, that is, the resource utilization rate of each node, is collected from each merged group, and then the loads of all nodes in the merged group are summed and divided by the number of nodes to obtain the average load of the nodes in the merged group. This average load value represents the overall load level of the nodes in the merged group, and provides a benchmark for the subsequent evaluation of load deviation. Subsequently, the average load of the nodes in each merged group is compared with the set load balancing standard to evaluate the degree of load deviation of each merged group. If the value of the average load of the nodes in a merged group deviating from the load balancing standard is greater than the preset load deviation, the merged group is marked as load abnormal. The abnormal load deviation represents that the resources of the merged group are allocated too much or too little, resulting in poor production efficiency or waste of resources. For the merged groups with abnormal load deviation identified, a reorganization operation will be performed. The goal of the reorganization is to redistribute the loads of these abnormal merged groups so that their loads are closer to the target load level. During the reorganization process, some nodes in the merged group with high load are allocated to the merged group with low load, or nodes are moved from the group with low load to balance the load levels of the two to reduce the load deviation. After the abnormal merge group is reorganized, the multi-group merge set will be updated to ensure the load balance and resource utilization efficiency of each merge group. The updated multi-group merge set will contain the adjusted merge groups, and the average node load of each merge group will be more in line with the system load standard. Ultimately, the load deviation of each merge group in the multi-group merge set will be controlled within a reasonable range, ensuring that the entire production system runs more balanced and efficient.

[0036] In summary, the capacity optimization device for multi-level production scheduling provided by the embodiment of the present application has the following technical effects: A combination constraint setting module 1 is used to obtain a multi-level production schedule and set multiple combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the number of combinations of the corresponding level production schedule; a node set determination module 2 is used to determine the node set of each level production schedule in the multi-level production schedule and output multiple groups of node sets; a continuity analysis module 3 is used to input the information of each group of node sets into a continuity analysis model and obtain a continuity analysis result according to the continuity analysis model; a node set combination module 4 is used to combine each group of node sets with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple groups of merged sets, wherein each level of production schedule corresponds to a group of merged sets; a scheduling balance scheduling module 5 is used to determine the capacity target, perform scheduling balance scheduling on the multiple groups of merged sets with the capacity target, and output a global balance strategy. Through the above steps, the technical problems of resource waste and low data processing efficiency caused by independent calculation of production tasks in complex production lines are solved, and the technical effect of reducing data processing volume, improving resource utilization and reducing production capacity waste is achieved by conducting continuity analysis on each production line and pre-combining the production lines as a whole.

[0037] Embodiment 2 is based on the same inventive concept as a multi-level production scheduling capacity optimization device in the above embodiment. Figure 2 As shown, the embodiment of the present application provides a method for optimizing the capacity of multi-level production scheduling, the method comprising: Acquire a multi-level production schedule, set multiple combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the number of combinations of the corresponding level production schedule; determine the node set of each level production schedule in the multi-level production schedule, and output multiple groups of node sets; input the information of each group of node sets into a continuity analysis model, and obtain continuity analysis results according to the continuity analysis model; use the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to combine each group of node sets to obtain multiple groups of merged sets, wherein each level production schedule corresponds to a group of merged sets; determine the capacity target, schedule the multiple groups of merged sets for balanced scheduling based on the capacity target, and output a global balanced strategy.

[0038] Further, obtaining the continuity analysis result according to the continuity analysis model includes: The continuity analysis model analyzes the information of each group of node sets, calculates the continuity characteristics between different nodes, and the continuity characteristics between nodes include the continuity between nodes, the switching efficiency between nodes, and the resource compatibility between nodes; constructs a feature matrix based on the continuity characteristics between different nodes; establishes a weight matrix corresponding to the feature matrix, and performs matrix calculation on the feature matrix based on the weight matrix to obtain the continuity analysis result.

[0039] Furthermore, each group of node sets is combined with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple combined sets, including: Obtain node pairs with a connectivity greater than a preset connectivity in the connectivity analysis results; combine the node pairs with a connectivity greater than the preset connectivity to obtain an initial merge group; record the current number of nodes in each initial merge group, and if the current number of nodes in each initial merge group is greater than or equal to the corresponding value in the multiple combination constraint factors, obtain a multiple-group merge set.

[0040] Furthermore, the method further comprises: If the current number of nodes in each initial merge group is less than the corresponding value in the multiple combination constraint factors, obtain a secondary combination instruction; recalculate the continuity analysis result corresponding to the initial merge group according to the secondary combination instruction; perform secondary combination on the initial merge group according to the recalculated continuity analysis result until the current number of nodes in each initial merge group is greater than or equal to the corresponding value in the multiple combination constraint factors.

[0041] Furthermore, the multi-group combined set is scheduled and balanced according to the production capacity target, and a global balancing strategy is output, including: Receive the capacity target, which includes a preset load variance; perform capacity load evaluation on each merged group in each merged set, and output the load evaluation data of each merged group; construct a balancing objective function based on the capacity target, call the balancing objective function to perform scheduling balancing response to the load evaluation data of each merged group, and output a global balancing strategy, wherein the global balancing strategy includes multiple balancing strategies corresponding to the multi-level production schedule.

[0042] Furthermore, multiple groups of merged sets are obtained, including: The average node load of each merged group in each merged set is obtained; according to the average node load of each merged group, abnormal merged groups with load deviation are identified, the abnormal merged groups are reorganized, and the multiple merged sets are updated.

[0043] Furthermore, multiple combined constraint factors are set for the multi-level production schedule, including: Collect historical node collaboration data sets for each level of production scheduling; extract the number of collaboration nodes from the historical node collaboration data sets for each level of production scheduling, and output a sample of the number of collaboration nodes; obtain the preferred number of collaboration nodes for each level of production scheduling based on the sample of the number of collaboration nodes, wherein the preferred number of collaboration nodes is the number with the highest frequency; set multiple combination constraint factors based on the preferred number of collaboration nodes for each level of production scheduling.

[0044] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0045] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to the selection of multiple elements individually or in whole. Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application intends to include these modifications and variations.

Claims

1. A capacity optimization device for multi-level production scheduling, characterized in that: The device comprises: A combination constraint setting module, used to obtain a multi-level production schedule, and set a plurality of combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the combination quantity of the corresponding level production schedule; A node set determination module, used to determine the node set of each level of production schedule in the multi-level production schedule, and output multiple groups of node sets; A continuity analysis module, used for inputting the information of each node set into a continuity analysis model, and obtaining a continuity analysis result according to the continuity analysis model; A node set combination module, used to combine each group of node sets with the multiple combination constraint factors as constraint conditions and the continuity analysis results as input variables to obtain multiple groups of combined sets, wherein each level of production scheduling corresponds to a group of combined sets; The scheduling and balancing module is used to determine the capacity target, perform scheduling and balancing on the multiple group combination sets according to the capacity target, and output a global balancing strategy.

2. The device according to claim 1, characterized in that The device for obtaining the continuity analysis result according to the continuity analysis model includes: The continuity analysis model analyzes the information of each set of nodes and calculates the continuity characteristics between different nodes, wherein the continuity characteristics include the continuity between nodes, the switching efficiency between nodes and the resource compatibility between nodes; Construct a feature matrix based on the connectivity characteristics between different nodes; A weight matrix corresponding to the feature matrix is ​​established, and matrix calculation is performed on the feature matrix according to the weight matrix to obtain a continuity analysis result.

3. The device according to claim 1, characterized in that The device is configured to combine each node set with the plurality of combined constraint factors as constraint conditions and the continuity analysis result as input variables to obtain a plurality of combined sets, and comprises: Obtaining node pairs with a connectivity greater than a preset connectivity in the connectivity analysis result; Combine the node pairs that are greater than the preset connectivity to obtain the initial merged group; The current number of nodes in each initial merge group is recorded, and if the current number of nodes in each initial merge group is greater than or equal to the corresponding value in the multiple combination constraint factors, a multiple group merge set is obtained.

4. The device according to claim 3, characterized in that If the current number of nodes in each initial merge group is less than the corresponding value in the multiple combination constraint factors, obtain a secondary combination instruction; According to the secondary combination instruction, recalculate the continuity analysis result corresponding to the initial merged group; The initial merged groups are recombined according to the recalculated continuity analysis results until the current number of nodes in each initial merged group is greater than or equal to the corresponding value of the multiple combination constraint factors.

5. The device according to claim 1, characterized in that The device performs scheduling and balancing on the multiple group merge sets according to the production capacity target and outputs a global balancing strategy, and includes: receiving the capacity target, wherein the capacity target includes a preset load variance; Perform capacity load evaluation on each merged group in each merged set, and output load evaluation data of each merged group; A balancing objective function is constructed based on the production capacity target, and the balancing objective function is called to perform a scheduling balancing response to the load assessment data of each merged group, and a global balancing strategy is output, wherein the global balancing strategy includes multiple balancing strategies corresponding to the multi-level production schedule.

6. The device according to claim 5, characterized in that Acquire multiple groups of combined sets, the device also includes: Obtaining the average load of nodes of each merge group in each merge set; According to the average load of the nodes of each merged group, an abnormal merged group with load deviation is identified, the abnormal merged group is reorganized, and the multiple merged sets are updated.

7. The device according to claim 1, characterized in that The device for setting a plurality of combined constraint factors for the multi-level production schedule comprises: Collect historical node collaboration data sets for each level of production scheduling; Extract the number of collaborative nodes from the historical node collaborative data set of each level of production scheduling, and output a sample of the number of collaborative nodes; According to the sample of the number of collaboration nodes, obtaining the preferred number of collaboration nodes for each level of production scheduling, wherein the preferred number of collaboration nodes is the number with the highest frequency; Multiple combination constraint factors are set based on the optimal number of collaborative nodes for each level of production scheduling.

8. A capacity optimization method for multi-level production scheduling, characterized in that: The method for optimizing the capacity of a multi-level production schedule is performed by a device for optimizing the capacity of a multi-level production schedule according to any one of claims 1 to 7, comprising: Acquire a multi-level production schedule, and set a plurality of combination constraint factors for the multi-level production schedule, wherein each combination constraint factor is used to constrain the combination quantity of the corresponding level production schedule; Determine a node set of each level of the multi-level production schedule, and output multiple groups of node sets; Inputting information of each group of node sets into a continuity analysis model, and obtaining continuity analysis results according to the continuity analysis model; Taking the multiple combined constraint factors as constraint conditions and the continuity analysis results as input variables, each group of node sets is combined to obtain multiple groups of combined sets, wherein each level of production scheduling corresponds to a group of combined sets; Determine a capacity target, perform scheduling and balancing on the multiple group merge sets based on the capacity target, and output a global balancing strategy.

Images