Clothing production line dynamic scheduling method and system based on digital twinning

Abstract

The invention relates to the technical field of electric digital data processing, in particular to a dynamic scheduling method and system for a garment production line based on digital twinning, and the method comprises the steps: receiving a plurality of garment orders sent by a customer, and according to the first description information of a garment process in the garment orders, generating a plurality of garment production lines; grouping the clothing orders in each preset time period to obtain a plurality of order groups; according to the first description information, generating production processes corresponding to the clothing process, and based on second description information for clothing materials in the clothing order, grouping the production processes to obtain a plurality of process groups; based on the parameters of the digital twin model corresponding to the production equipment, clustering operation is carried out on the process groups in the same order group to obtain a plurality of clusters, and the execution sequence of the process groups in the order group is determined according to the clustering distance between the process groups in the clusters and the production equipment; and sending the execution sequence to an edge end corresponding to the clustering cluster.

Description

A method and system for dynamic scheduling of garment production lines based on digital twins Technical Field

[0001] This application relates to the field of digital data processing technology, and in particular to a dynamic scheduling method and system for garment production lines based on digital twins. Background Technology

[0002] In today's fast-paced and diversified market, consumer demand is showing an unprecedented trend of diversification. With economic development and improved living standards, people are no longer satisfied with uniform clothing but pursue personalized and distinctive garments. This rapid change in market demand has prompted the garment industry to shift towards a more flexible production model. Among these, the "multi-variety, small-batch, customized" garment production model has emerged and demonstrated significant advantages.

[0003] However, for garment production lines, the "multi-variety, small-batch, customized" production model typically involves frequent switching of production materials and / or equipment parameters, thereby reducing the unit capacity of the garment production line and affecting delivery speed. Furthermore, frequent switching of production materials or equipment parameters can cause fluctuations in garment quality, leading to an increase in the defect rate and raising the probability of equipment malfunctions. Summary of the Invention

[0004] This application provides a method and system for dynamic scheduling of garment production lines based on digital twins, in order to avoid the negative impact caused by frequent switching of production materials or equipment parameters.

[0005] In a first aspect, this application provides a dynamic scheduling method for a garment production line based on digital twins, used in the cloud. The method includes: receiving multiple garment orders sent by customers, and grouping the garment orders within each preset time period according to a first description of the garment process in the garment orders to obtain multiple order groups; generating production processes corresponding to the garment process according to the first description information, and grouping the same production process corresponding to different garment orders in the same order group according to a second description of the garment material in the garment orders to obtain multiple process groups; performing clustering operations on the process groups in the same order group based on the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters, each cluster containing one production equipment and multiple process groups, and determining the execution order of the process groups in the order group according to the clustering distance between the process groups and the production equipment in the cluster; sending the execution order to an edge terminal corresponding to the cluster, so that the edge terminal adjusts the execution order according to the production status of the production equipment, and sends a production scheduling instruction matching the adjusted execution order to each production equipment.

[0006] Furthermore, the step of grouping garment orders within each preset time period according to the first description information of garment process in the garment order to obtain multiple order groups with different production sequences includes: determining the process duration corresponding to the garment process according to the third description information of garment design and the first description information in the garment order; constructing a process curve corresponding to each garment order with the process duration of each garment order within the preset time period as the horizontal axis and the parameters corresponding to the garment process in the first description information as the vertical axis; and grouping the garment orders within the preset time period according to the matching data between the process curves corresponding to different garment orders to obtain multiple order groups.

[0007] Furthermore, the matching data includes first matching data and second matching data. The step of grouping garment orders within the preset time period based on the matching data between the process curves corresponding to different garment orders to obtain multiple order groups includes: performing a first grouping operation on garment orders within the same preset time period based on the first matching data to obtain multiple initial order groups, where the first matching data represents the number of positive and negative changes in the slope of the process curve; and performing a second grouping operation on the initial order groups based on the second matching data to obtain multiple order groups, where the second matching data represents the degree of difference in process time corresponding to the same garment process.

[0008] Furthermore, the step of clustering process groups within the same order group based on the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters includes: reading the historical production records of the production equipment in the digital twin model, and determining the target process based on the historical production records, wherein the target process is the process with the lowest probability of disturbance events occurring when the production equipment produces different production processes, and the historical production records are sent by the edge device; determining the target parameters corresponding to the production equipment based on the parameters of the target process, and determining the mean parameter corresponding to the process group based on the parameters of each production process in the process group; and performing clustering operations on the process groups within the same order group based on the target parameters and the mean parameters to obtain multiple clusters.

[0009] Furthermore, determining the target parameters corresponding to the production equipment based on the parameters of the target process, and determining the mean parameter corresponding to the process group based on the parameters of each production process in the process group, includes: encoding the parameters of the target process to obtain a first representation vector; averaging the parameters of each process in the process group and encoding them to obtain a second representation vector; concatenating the first representation vectors corresponding to the target process to obtain the target parameters; and concatenating the second representation vectors corresponding to the process group to obtain the mean parameter.

[0010] Furthermore, determining the execution order of process groups in the order group based on the clustering distance between the process group and the production equipment in the cluster includes: sorting the process groups according to the clustering distance between the process group and the production equipment in the cluster to obtain a sorting result; and determining the execution order of different process groups in the same order group based on the sorting result.

[0011] Secondly, this application provides a dynamic scheduling method for a garment production line based on digital twins, used at the edge, comprising: receiving an execution order sent from the cloud; the cloud determining the execution order of process groups in an order group based on the clustering distance between process groups and production equipment in a cluster; the cloud performing a clustering operation on process groups in the same order group based on parameters of the digital twin model corresponding to each production equipment, obtaining multiple clusters, each cluster containing one production equipment and multiple process groups; the cloud grouping the same production process corresponding to different garment orders in the same order group based on a second description information of the garment material in the garment order, obtaining multiple process groups; the cloud generating a production process corresponding to the garment process based on a first description information; the cloud grouping garment orders within each preset time period based on the first description information of the garment process in the garment order, obtaining multiple order groups; multiple garment orders being sent to the cloud by the customer; adjusting the execution order according to the production status of the production equipment, and sending a production scheduling instruction matching the adjusted execution order to each production equipment.

[0012] Furthermore, adjusting the execution order based on the production status of the production equipment includes: taking the total time required to complete each production process in the process group as a first time, and based on the first time, determining the total time required to complete each production process in the cluster as a second time; when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, obtaining an adjustment coefficient corresponding to the cluster, wherein the length of the second time of the cluster is directly proportional to the adjustment coefficient of the cluster; using the adjustment coefficient to adjust the cluster distance between the process group and the production equipment in the cluster, so that the difference between the second times corresponding to the same type of production equipment is less than or equal to a preset value; and determining the execution order of the process groups in the order group based on the adjusted cluster distance in the cluster.

[0013] Furthermore, when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, obtaining the adjustment coefficient corresponding to the clusters includes: when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, determining the second time sum, where the second time sum is the sum of the second times of each cluster; and determining the adjustment coefficient based on the ratio of the second time of the clusters to the second time sum.

[0014] Thirdly, this application provides a dynamic scheduling system for a garment production line based on digital twins, including a cloud and an edge terminal; the cloud is used to: receive multiple garment orders sent by customers, and group the garment orders within each preset time period according to the first description information of the garment process in the garment orders to obtain multiple order groups with different production sequences; generate production processes corresponding to the garment process according to the first description information, and group the same production process corresponding to different garment orders in the same order group according to the second description information of the garment material in the garment orders to obtain multiple process groups; perform clustering operations on the process groups in the same order group based on the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters, each cluster containing one production equipment and multiple process groups, and determine the execution sequence of the process groups in the order group according to the clustering distance between the process groups and the production equipment in the cluster; send the execution sequence to the edge terminal corresponding to the cluster; the edge terminal is used to: receive the execution sequence sent by the cloud; adjust the execution sequence according to the production status of the production equipment, and send production scheduling instructions matching the adjusted execution sequence to each production equipment.

[0015] The present invention has at least the following technical effects: In the embodiments of this specification, garment orders are first grouped into multiple order groups based on the first description information (first classification); then, the production processes corresponding to the garment orders are grouped into multiple process groups based on the second description information (second classification); finally, the process groups are clustered based on the parameters of the digital twin model (third classification), and the execution order between process groups is determined based on the clustering. This process, through the combination of three classifications, not only achieves the sequential execution of garment orders with the same process and the sequential execution of the same production processes, but also achieves hierarchical execution between different processes, greatly reducing the frequency of switching production materials or equipment parameters. While improving the unit capacity and supply speed of the garment production line, it can also avoid fluctuations in garment quality, reduce the defect rate, and reduce the probability of production equipment failure. Attached Figure Description

[0016] 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.

[0017] Figure 1 illustrates a flowchart of a dynamic scheduling method for a garment production line based on digital twins; Figure 2 illustrates a flowchart of a dynamic scheduling method for a garment production line based on digital twins; Figure 3 illustrates a structural diagram of a dynamic scheduling system for a garment production line based on digital twins.

[0018] 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

[0019] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0020] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0021] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0022] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0023] Figure 1 shows a flowchart of a dynamic scheduling method for a garment production line based on digital twins according to an embodiment of the present disclosure. This method can be applied to a dynamic scheduling device for a garment production line based on digital twins. The dynamic scheduling device for a garment production line based on digital twins can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.

[0024] In some possible implementations, this digital twin-based dynamic scheduling method for garment production lines can be implemented by having the processor call computer-readable instructions stored in memory.

[0025] As shown in Figure 1, the dynamic scheduling method for a garment production line based on digital twins, used in the cloud, includes: step S11, receiving multiple garment orders sent by customers, and grouping the garment orders within each preset time period according to the first description information of the garment process in the garment orders to obtain multiple order groups.

[0026] The garment order may include various descriptive information: first descriptive information, second descriptive information, and third descriptive information. The first descriptive information describes the garment's construction process (e.g., the positional relationship between garment components, stitch types, etc.), the second descriptive information describes the garment's material (e.g., the material and color of garment components, etc.), and the third descriptive information describes the garment's design (e.g., the overall structure and dimensions of the garment, etc.). This specification does not specify a particular format for the descriptive information; it can be determined based on the actual situation. For example, the first descriptive information may be size data, processing details, etc.; the second descriptive information may be material data, color data, etc.; and the third descriptive information may be design documents such as flat drawings or three-dimensional drawings.

[0027] Garment manufacturing process generally describes the overall process of making a garment. Therefore, upon receiving multiple garment orders from a customer, the orders can be initially categorized based on the initial description of the garment manufacturing process. Specifically, a time period can be set to indicate the delivery time of each garment order. Then, based on the initial description of each garment order within that time period, the orders are grouped into multiple order groups, where the garment orders within each group have the same or similar manufacturing processes. This allows the garment orders to be sorted according to the resulting order groups and processed sequentially.

[0028] Step S12: Based on the first description information, generate the production process corresponding to the garment process, and based on the second description information of the garment material in the garment order, group the same production process corresponding to different garment orders in the same order group to obtain multiple process groups.

[0029] In this context, garment manufacturing processes refer to the streamlined steps in garment production (such as cutting pieces and sewing), each step being considered a process. Specifically, a process database or knowledge graph can be established for garment manufacturing processes. In this database, each process corresponds to at least one process. Thus, after obtaining the initial description information, the corresponding process can be directly retrieved from the process database based on that information, thereby obtaining the overall process corresponding to the garment order. The method for determining production processes based on the initial description information of the garment manufacturing process can be determined according to actual circumstances, and this specification does not impose any restrictions on it.

[0030] Typically, a single production machine can complete one or more processes. After the initial classification mentioned above, a second classification can be performed on the generated garment processes. This second classification primarily refers to the classification within the same production process. Specifically, for the same production process, different garment orders may use different garment materials within that process, causing the garment equipment to switch between materials. Therefore, the second description information regarding the garment material can be combined with the production process to group the production processes, resulting in multiple process groups. Each process group contains the same production process for multiple garment orders, and these production processes share the same second description information.

[0031] Step S13: Based on the parameters of the digital twin model corresponding to the production equipment, perform clustering operations on the process groups in the same order group to obtain multiple clusters. Each cluster contains one production equipment and multiple process groups. The execution order of the process groups in the order group is determined according to the clustering distance between the process groups and the production equipment in the cluster.

[0032] Step S14: The execution order is sent to the edge terminal corresponding to the cluster, so that the edge terminal adjusts the execution order according to the production status of the production equipment and sends a production scheduling instruction matching the adjusted execution order to each of the production equipment.

[0033] Digital twin technology, as a crucial tool for achieving virtual-physical collaboration in production lines, possesses real-time perception, predictive decision-making, and optimized control capabilities, significantly improving scheduling response efficiency and precisely controlling energy consumption distribution. Therefore, digital twin technology can be used to schedule and manage garment production lines. Specifically, a digital twin garment production line can be built in the cloud for the physical garment production line. This digital twin production line includes digital twin models corresponding to the production equipment. Through edge devices, information such as equipment process parameters and product quality data can be sent to the cloud, and scheduling data from the cloud (such as order execution order) can also be sent to the edge devices. An edge device can be set up for each type of production equipment, so that each edge device is responsible for scheduling the same type of production process.

[0034] The parameters of a cloud-based digital twin model can include the process parameters and product quality parameters of the corresponding production equipment. These parameters can indicate which production processes the equipment is better suited for in garment manufacturing. In one example, the cloud can cluster process groups within the same order group based on the parameters of each production equipment on the production line, determining which production equipment is most suitable to assign each process group within that order group. This clustering process can be viewed as a third classification. The cluster centers can be the production equipment, and the cluster distance between the process group and the cluster center can be used to determine the execution order.

[0035] In one possible implementation, determining the execution order of process groups in the order group based on the clustering distance between the process group and the production equipment in the cluster includes: sorting the process groups according to the clustering distance between the process group and the production equipment in the cluster to obtain a sorting result; and determining the execution order of different process groups in the same order group based on the sorting result.

[0036] Specifically, based on the cluster distances within the same cluster, the process groups corresponding to those cluster distances can be sorted to obtain a sorting result. This sorting result is a sorting sequence, which can be from larger cluster distances to smaller cluster distances, or from smaller cluster distances to smaller cluster distances. This sorting result represents the execution order of the process groups.

[0037] Since the execution result is obtained based on the sorting of cluster distance, and the cluster distance reflects the matching degree between the process group and the production equipment, the above process realizes the hierarchical execution of the process group (process groups with similar cluster distances can be regarded as the same level). Since there are fewer material changes and parameter adjustments of the production equipment when switching between process groups at adjacent levels, this process reduces the negative impact of frequent switching of production materials or equipment parameters to a certain extent.

[0038] Since each production process is completed by at least one production device, and each cluster corresponds to one production device, each edge device corresponds to at least one cluster. Specifically, the process groups can first be clustered in the cloud based on digital twin technology to obtain the initial execution order of the process groups. Then, at the edge device, the execution order is adjusted according to the production status of the production devices (such as whether there is a disturbance, the duration of the disturbance, etc.). Finally, the edge device sends production scheduling instructions to the managed production devices according to the adjusted execution order, enabling the production devices to fulfill garment orders. This process sets the execution order determination process with heavier computational tasks in the cloud with stronger computing power, and sets the execution order adjustment process with lighter computational tasks in the edge device with weaker computing power, realizing edge-cloud collaboration based on digital twins.

[0039] A garment order corresponds to multiple production processes ordered in sequence, with each process belonging to a different process group. For any process group (which can be called the target process group), the edge endpoint can determine the execution order of each production process within the target process group based on the coordination of production processes in the process groups preceding and following the target process group (where "before" and "after" are used to indicate the processing order). This specification does not require a specific process for determining the execution order of production processes within a process group; it can be obtained by modeling a conventional shop floor scheduling problem.

[0040] In the embodiments of this specification, garment orders are first grouped into multiple order groups based on the first description information (first classification). Then, based on the second description information, the production processes corresponding to the garment orders are grouped into multiple process groups (second classification). Finally, the process groups are clustered based on the parameters of the digital twin model (third classification). Based on the clustering, the execution order between the process groups is determined. This process, through the combination of three classifications, not only achieves the sequential execution of garment orders with the same process and the sequential execution of the same production processes, but also achieves hierarchical execution between different processes. This significantly reduces the frequency of switching production materials or equipment parameters, improving the unit capacity and supply speed of the garment production line while also avoiding fluctuations in garment quality, reducing the defect rate, and lowering the probability of production equipment failure.

[0041] Due to its intuitive and visual nature, line graphs are widely used in data analysis. In one possible implementation, the step of grouping garment orders within each preset time period based on the first description information regarding garment technology in the garment orders to obtain multiple order groups with different production sequences includes: determining the process duration corresponding to the garment technology based on the third description information regarding garment design and the first description information in the garment orders; constructing a process curve corresponding to each garment order with the process duration of each garment order within the preset time period as the horizontal axis and the parameters corresponding to the garment technology in the first description information as the vertical axis; and grouping the garment orders within the preset time period based on the matching data between the process curves corresponding to different garment orders to obtain multiple order groups.

[0042] The first description information usually includes various process-related parameters, which can indicate the time range for completing the garment process; the third description information usually includes data such as the dimensions of various parts of the garment (such as stitch size, pattern size, etc.) and processing difficulty (such as the processing difficulty of different stitches such as chain stitch and imitation hand stitch is different). These parameters can further refine the time required to complete the garment process (i.e., the specific value within the aforementioned time range). Therefore, the process duration corresponding to the garment process can be determined based on the first and third description information.

[0043] In one example, for garment orders within the same preset time period, a process curve can be constructed with process duration as the x-axis and garment process parameters as the y-axis. Specifically, when determining the x-axis, process duration can be used directly, or the x-axis of the first garment process can be its process duration, with the x-axis of subsequent garment processes adding the process duration of the preceding garment processes. When determining the y-axis, to differentiate between each garment process, the parameters of each garment process can be normalized to different values. Furthermore, in the process curve, the order of the garment processes should be consistent with the execution order of the garment processes during garment processing.

[0044] After obtaining the process curves for different garment orders, the garment orders can be grouped according to the differences between the process curves to obtain order groups.

[0045] In one possible implementation, the matching data includes first matching data and second matching data. The step of grouping clothing orders within the preset time period according to the matching data between the process curves corresponding to different clothing orders to obtain multiple order groups includes: performing a first grouping operation on clothing orders within the same preset time period according to the first matching data to obtain multiple initial order groups; and performing a second grouping operation on the initial order groups based on the second matching data to obtain multiple order groups.

[0046] The first matching data represents the number of positive and negative changes in the slope of the process curve. Clearly, the first matching data reflects the number of peaks and troughs in the process curve, which is directly related to the number of garment processes. Different garment orders may have different numbers of garment processes. For example, some garments may not have printing or embroidery, while others may. For the former, the garment process will include printing or embroidery. Therefore, based on the first matching data, a first grouping operation can be performed on the orders, grouping garment orders with the same garment processes into one category.

[0047] The second matching data is used to represent the degree of difference in process time corresponding to the same garment technology. The process time for different garment technologies usually varies greatly. For example, the process time for batik is calculated in hours, while the process time for printing is calculated in minutes. After simply grouping garment orders with the same garment technology into one category, in order to avoid situations where the number of garment technologies is the same but the content of the garment technologies is inconsistent (e.g., two garment orders both include the four garment technologies a, b, c, and d, but in one garment order c is batik and in the other garment order c is printing), a second grouping operation can be performed based on the second matching data to obtain the final order group. Within the same order group, the degree of difference in process time between garment technologies is small.

[0048] The above process, through two-level grouping (i.e., the first grouping operation and the second grouping operation), achieves precise division of order groups, which in turn helps to improve the accuracy of subsequent production process grouping and other operations.

[0049] As production equipment is continuously used and debugged, the product quality varies when each piece of equipment performs different production processes. In one possible implementation, the process group within the same order group is clustered based on the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters. This includes: reading the historical production records of the production equipment in the digital twin model and determining the target process based on the historical production records, which are sent by the edge device; determining the target parameters corresponding to the production equipment based on the parameters of the target process; determining the mean parameter corresponding to the process group based on the parameters of each production process in the process group; and performing a clustering operation on the process group within the same order group based on the target parameters and the mean parameter to obtain multiple clusters.

[0050] The target process is the process with the lowest probability of occurrence of disturbance events when the production equipment performs different production processes. In one example, since the target process has the lowest probability of occurrence of disturbance events, the parameters corresponding to the target process can be used to represent the production equipment, and the parameters corresponding to the production processes can be used to represent the process group, thereby performing clustering operations. The parameters corresponding to the target process can be matched one-to-one with the parameters corresponding to the production processes in the process group.

[0051] Specifically, based on a preset parameter order, data corresponding to the production equipment (i.e., target parameters) and data corresponding to the production process (i.e., mean parameters) can be obtained during clustering operations. When the number of parameters corresponding to the target process and the production process is large or the parameter dimensions are inconsistent, the parameters can be encoded to unify the parameter dimensions during clustering operations, thereby improving the speed of clustering operations. In one possible implementation, determining the target parameters corresponding to the production equipment based on the parameters of the target process, and determining the mean parameters corresponding to the process group based on the parameters of each production process in the process group, includes: encoding the parameters of the target process to obtain a first representation vector; averaging the parameters of each process in the process group and encoding them to obtain a second representation vector; concatenating the first representation vectors corresponding to the target process to obtain the target parameters; and concatenating the second representation vectors corresponding to the process group to obtain the mean parameters.

[0052] Specifically, the parameters corresponding to the target process and the production process can be encoded first to obtain a first representation vector and a second representation vector. The first and second representation vectors have the same dimension. Then, based on the preset parameter order, the representation vectors corresponding to the target process and the production process can be concatenated. The concatenated vectors are used as the target parameter and the mean parameter. Then, the target parameter and the mean parameter in vector form can be used to perform clustering operations to obtain clusters.

[0053] After the production equipment at the edge completes the garment order in the order group, it can send the quality data of the garment order to the cloud, so that the cloud can update the parameters of the digital twin model corresponding to the production equipment based on the quality data.

[0054] Figure 2 shows a flowchart of a dynamic scheduling method for a garment production line based on digital twins according to an embodiment of the present disclosure. This method can be applied to a dynamic scheduling device for a garment production line based on digital twins. The dynamic scheduling device for a garment production line based on digital twins can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, wearable device, etc.

[0055] In some possible implementations, this digital twin-based dynamic scheduling method for garment production lines can be implemented by having the processor call computer-readable instructions stored in memory.

[0056] As shown in Figure 2, the dynamic scheduling method for a garment production line based on digital twins, used at the edge, may include: Step S21, receiving the execution order sent by the cloud, wherein the cloud determines the execution order of the process groups in the order group based on the clustering distance between the process groups and production equipment in the cluster; the cloud performs clustering operation on the process groups in the same order group based on the parameters of the digital twin model corresponding to each production equipment, obtaining multiple clusters, each cluster containing one production equipment and multiple process groups; the cloud groups the same production process corresponding to different garment orders in the same order group based on the second description information of the garment material in the garment order, obtaining multiple process groups; the cloud generates the production process corresponding to the garment process based on the first description information; the cloud groups the garment orders within each preset time period based on the first description information of the garment process in the garment order, obtaining multiple order groups; multiple garment orders are sent to the cloud by the customer; Step S22, adjusting the execution order according to the production status of the production equipment, and sending a production scheduling instruction matching the adjusted execution order to each production equipment.

[0057] In the embodiments of this specification, garment orders are first grouped into multiple order groups based on the first description information (first classification). Then, based on the second description information, the production processes corresponding to the garment orders are grouped into multiple process groups (second classification). Finally, the process groups are clustered based on the parameters of the digital twin model (third classification). Based on the clustering, the execution order between the process groups is determined. This process, through the combination of three classifications, not only achieves the sequential execution of garment orders with the same process and the sequential execution of the same production processes, but also achieves hierarchical execution between different processes. This significantly reduces the frequency of switching production materials or equipment parameters, improving the unit capacity and supply speed of the garment production line while also avoiding fluctuations in garment quality, reducing the defect rate, and lowering the probability of production equipment failure.

[0058] In a cluster, the cluster center is the production equipment. After the clustering operation is completed, it is very likely that the number of production processes corresponding to a production equipment will be different. This may lead to different production equipment taking different times to complete the assigned production processes. This will cause some production equipment to be idle for too long while others work for too long. When the workload of each production equipment is too unbalanced, it will reduce the processing efficiency of garment orders and cause serious damage to some equipment. In one possible implementation, adjusting the execution order based on the production status of the production equipment includes: taking the sum of the time required to complete each production process in the process group as a first time, and based on the first time, determining the sum of the time required to complete each production process in the cluster as a second time; when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, obtaining an adjustment coefficient corresponding to the cluster, wherein the length of the second time of the cluster is directly proportional to the adjustment coefficient of the cluster; using the adjustment coefficient to adjust the cluster distance between the process group and the production equipment in the cluster, so that the difference between the second times corresponding to the same type of production equipment is less than or equal to a preset value; and determining the execution order of the process groups in the order group based on the adjusted cluster distance in the cluster.

[0059] Specifically, the first time required to complete each production process in each process group can be calculated first. Then, the first times of each process group in the cluster are summed to obtain the second time. At this point, the difference in the second time between specific clusters can be compared. These specific clusters are production equipment of the same or similar type that can complete the same or similar production processes. If the difference is greater than a preset value, it indicates that some production equipment may be idle. The cluster distance in the cluster can be adjusted to update the cluster. Then, the new execution order can be determined by the cluster distance corresponding to the updated cluster, thus achieving a balance between the workload of different production equipment.

[0060] The adjustment coefficient corresponding to each cluster can be directly proportional to the length of the second time interval. That is, the longer the second time interval, the larger the adjustment coefficient, and consequently, the larger the cluster distance obtained based on the adjustment coefficient. This means that process groups within a cluster may be reassigned to other clusters with shorter second times during the adjustment process. Conversely, the shorter the second time interval, the smaller the adjustment coefficient, and process groups from other clusters may be adjusted to those clusters with shorter second times. The adjustment coefficient can be obtained empirically or based on the second time interval. In one possible implementation, obtaining the adjustment coefficient for a cluster when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value includes: determining the total second time interval when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value; and determining the adjustment coefficient based on the ratio of the second time interval of the cluster to the total second time interval.

[0061] For multiple clusters of production equipment of the same type, the sum of the second time intervals for these clusters can be calculated. The second time interval for each cluster can be compared to the sum of the second times, and this ratio can be used as an adjustment coefficient. Since clusters with longer second times correspond to larger ratios, and clusters with shorter second times correspond to smaller ratios, this adjustment coefficient simultaneously adjusts both clusters with longer and shorter second times, enabling rapid updates to the clusters.

[0062] Figure 3 presents a block diagram of a dynamic scheduling system for a garment production line based on digital twins. As shown in Figure 3, this dynamic scheduling system for a garment production line based on digital twins includes cloud and edge terminals. The workflow of this dynamic scheduling system for a garment production line based on digital twins can be referred to the aforementioned embodiments, and will not be repeated here.

[0063] This invention is now complete.

[0064] In summary, in the embodiments of this specification, garment orders are first grouped into multiple order groups based on the first description information (first classification). Then, the production processes corresponding to the garment orders are grouped into multiple process groups based on the second description information (second classification). Finally, the process groups are clustered based on the parameters of the digital twin model (third classification). Based on the clustering, the execution order between the process groups is determined. This process, through the combination of three classifications, not only achieves the sequential execution of garment orders with the same process and the sequential execution of the same production processes, but also achieves hierarchical execution between different processes. This significantly reduces the frequency of switching production materials or equipment parameters, improving the unit capacity and supply speed of the garment production line while avoiding fluctuations in garment quality, reducing the defect rate, and lowering the probability of production equipment failure.

[0065] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A dynamic scheduling method for a garment production line based on digital twins, characterized in that, For use in the cloud, the method includes: receiving multiple clothing orders sent by customers, and grouping the clothing orders within each preset time period according to the first description information of clothing process in the clothing orders to obtain multiple order groups; generating production processes corresponding to the clothing process according to the first description information, and grouping the same production process corresponding to different clothing orders in the same order group according to the second description information of clothing material in the clothing orders to obtain multiple process groups; performing clustering operations on the process groups in the same order group according to the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters, each cluster containing one production equipment and multiple process groups, and determining the execution order of the process groups in the order group according to the clustering distance between the process groups and the production equipment in the cluster; sending the execution order to the edge terminal corresponding to the cluster, so that the edge terminal adjusts the execution order according to the production status of the production equipment, and sends a production scheduling instruction matching the adjusted execution order to each production equipment.

2. The method according to claim 1, characterized in that, The step of grouping garment orders within each preset time period according to the first description information of garment process in the garment order to obtain multiple order groups with different production sequences includes: determining the process duration corresponding to the garment process according to the third description information of garment design and the first description information in the garment order; constructing a process curve corresponding to each garment order with the process duration of each garment order within the preset time period as the horizontal axis and the parameters corresponding to the garment process in the first description information as the vertical axis; and grouping the garment orders within the preset time period according to the matching data between the process curves corresponding to different garment orders to obtain multiple order groups.

3. The method according to claim 2, characterized in that, The matching data includes first matching data and second matching data. The step of grouping garment orders within the preset time period based on the matching data between the process curves corresponding to different garment orders to obtain multiple order groups includes: performing a first grouping operation on garment orders within the same preset time period based on the first matching data to obtain multiple initial order groups, where the first matching data represents the number of positive and negative changes in the slope of the process curve; and performing a second grouping operation on the initial order groups based on the second matching data to obtain multiple order groups, where the second matching data represents the degree of difference in process time corresponding to the same garment process.

4. The method according to claim 1, characterized in that, The method of clustering process groups within the same order group based on parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters includes: reading historical production records of the production equipment in the digital twin model and determining a target process based on the historical production records, wherein the target process is the process with the lowest probability of disturbance events occurring when the production equipment produces different production processes, and the historical production records are sent by the edge device; determining target parameters corresponding to the production equipment based on the parameters of the target process; determining the mean parameter corresponding to the process group based on the parameters of each production process in the process group; and performing clustering operations on the process groups within the same order group based on the target parameters and the mean parameters to obtain multiple clusters.

5. The method according to claim 4, characterized in that, The step of determining the target parameters corresponding to the production equipment based on the parameters of the target process, and determining the mean parameter corresponding to the process group based on the parameters of each production process in the process group, includes: encoding the parameters of the target process to obtain a first representation vector; averaging the parameters of each process in the process group and encoding them to obtain a second representation vector; concatenating the first representation vectors corresponding to the target process to obtain the target parameters; and concatenating the second representation vectors corresponding to the process group to obtain the mean parameter.

6. The method according to claim 1, characterized in that, The step of determining the execution order of process groups in the order group based on the clustering distance between the process groups and production equipment in the cluster includes: sorting the process groups according to the clustering distance between the process groups and production equipment in the cluster to obtain a sorting result; and determining the execution order of different process groups in the same order group based on the sorting result.

7. A dynamic scheduling method for a garment production line based on digital twins, characterized in that, For edge computing, the system includes: receiving an execution order sent from the cloud; the cloud determining the execution order of process groups in an order group based on the clustering distance between process groups and production equipment in a cluster; the cloud performing clustering operations on process groups in the same order group based on parameters of the digital twin model corresponding to each production equipment to obtain multiple clusters, each cluster containing one production equipment and multiple process groups; the cloud grouping the same production process corresponding to different clothing orders in the same order group based on the second description information of clothing material in the clothing order to obtain multiple process groups; the cloud generating production processes corresponding to clothing processes based on the first description information; the cloud grouping clothing orders within each preset time period based on the first description information of clothing processes in the clothing order to obtain multiple order groups; multiple clothing orders being sent to the cloud by customers; adjusting the execution order according to the production status of the production equipment, and sending production scheduling instructions matching the adjusted execution order to each production equipment.

8. The method according to claim 7, characterized in that, The step of adjusting the execution order based on the production status of the production equipment includes: taking the total time required to complete each production process in the process group as a first time, and based on the first time, determining the total time required to complete each production process in the cluster as a second time; when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, obtaining an adjustment coefficient corresponding to the cluster, wherein the length of the second time of the cluster is directly proportional to the adjustment coefficient of the cluster; using the adjustment coefficient to adjust the cluster distance between the process group and the production equipment in the cluster, so that the difference between the second times corresponding to the same type of production equipment is less than or equal to a preset value; and determining the execution order of the process groups in the order group based on the adjusted cluster distance in the cluster.

9. The method according to claim 8, characterized in that, When the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, the adjustment coefficient corresponding to the cluster is obtained, which includes: when the production equipment corresponding to different clusters is of the same type and the difference between the second times of the clusters exceeds a preset value, determining the second time sum, wherein the second time sum is the sum of the second times of each cluster; and determining the adjustment coefficient based on the ratio of the second time of the cluster to the second time sum.

10. A dynamic scheduling system for a garment production line based on digital twins, comprising a cloud and an edge terminal; the cloud is used for: receiving multiple garment orders sent by customers, and grouping garment orders within each preset time period according to a first description of the garment process in the garment orders to obtain multiple order groups with different production sequences; generating production processes corresponding to the garment process according to the first description information, and grouping the same production process corresponding to different garment orders in the same order group according to a second description of the garment material in the garment orders to obtain multiple process groups; performing clustering operations on the process groups in the same order group based on the parameters of the digital twin model corresponding to the production equipment to obtain multiple clusters, each cluster containing one production equipment and multiple process groups, and determining the execution sequence of the process groups in the order group according to the clustering distance between the process groups and the production equipment in the cluster; sending the execution sequence to the edge terminal corresponding to the cluster; the edge terminal is used for: receiving the execution sequence sent by the cloud; adjusting the execution sequence according to the production status of the production equipment, and sending production scheduling instructions matching the adjusted execution sequence to each of the production equipment.

Images