A scheduling method and device for a clothing production line

Abstract

The present invention relates to the field of production line scheduling, and more specifically to a scheduling method and device for a clothing production line, comprising: obtaining multiple clothing orders, assigning identical clothing processing tasks in each clothing order to a corresponding target edge computing node, wherein the target edge computing node is used to sequentially process identical clothing processing tasks; if the processing time of the target edge computing node exceeds a preset time, obtaining the remaining clothing processing tasks corresponding to the target edge computing node; obtaining the allocation relationship value between the remaining clothing processing tasks and other target edge computing nodes; and determining the optimal target edge computing node for the remaining clothing processing tasks based on the allocation relationship value. By detecting the computing delay of the edge computing node and based on the allocation relationship value, the computing power of all edge computing nodes in the production system is automatically rescheduled, thereby avoiding the problems of idle computing power and excessive load on individual devices existing in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of production line scheduling, and in particular to a scheduling method and device for a clothing production line. Background Art

[0002] Many technologies are currently available on the market to meet this need, among which edge computing-based data analysis and visualization technologies are common. This technology installs edge computing devices and data sensors at every processing step on every production line, aggregating the processed analysis results to a central control device for visualization. However, for apparel production, this process differs significantly from other industries. This is because orders may involve the production of multiple garment types, each requiring different processing techniques, each of which is completed by different production lines. Typically, different production lines are responsible for different production processes. Therefore, in actual production, the workload of different production lines varies. When the load on a production line deviates significantly, the computing power of some edge computing devices may be idle. Summary of the Invention

[0003] In order to solve the above-mentioned problems in the prior art, the present invention provides a scheduling method and device for a clothing production line, which solves the problem that the computing power of some edge computing devices may be idle when there is a serious deviation in the load of the production line.

[0004] According to one aspect of the present invention, a scheduling method for a garment production line is provided, comprising:

[0005] Get multiple clothing orders;

[0006] Assigning identical garment processing tasks in each garment order to corresponding target edge computing nodes; wherein the target edge computing nodes are used to sequentially process identical garment processing tasks;

[0007] If the processing time of the target edge computing node exceeds the preset time, the remaining clothing processing tasks corresponding to the target edge computing node are obtained; wherein the remaining clothing processing tasks are tasks that the target edge computing node has not completed within the processing time;

[0008] Obtaining the distribution relationship value between the remaining clothing processing tasks and other target edge computing nodes;

[0009] Based on the allocation relationship value, an optimal target edge computing node for allocating the remaining clothing processing tasks is determined.

[0010] In one embodiment, obtaining the allocation relationship value between the remaining clothing processing tasks and other target edge computing nodes includes:

[0011] Determining the execution time corresponding to the remaining clothing processing tasks;

[0012] Determine the computing capacity values ​​of the remaining clothing processing tasks allocated to other target edge computing nodes;

[0013] The allocation relationship value between the remaining clothing processing tasks and the other target edge computing nodes is determined according to the execution time corresponding to the remaining clothing processing tasks and the computing capacity value of the remaining clothing processing tasks allocated to the other target edge computing nodes.

[0014] In one embodiment, determining the optimal target edge computing node for allocating the remaining clothing processing tasks based on the allocation relationship value includes:

[0015] Obtaining the current delay time of the target edge computing node whose processing time exceeds the preset time;

[0016] Get the average latency of multiple target edge computing nodes;

[0017] The optimal target edge computing node for assigning the remaining clothing processing tasks is determined according to the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the multiple target edge computing nodes, and the allocation relationship value.

[0018] In one embodiment, determining the optimal target edge computing node to which the remaining clothing processing tasks are assigned based on the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the multiple target edge computing nodes, and the allocation relationship value includes:

[0019] updating the allocation cost according to the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the plurality of target edge computing nodes, and the allocation relationship value;

[0020] If the allocation cost is less than a preset cost threshold, the optimal target edge computing node for the allocation of the remaining clothing processing tasks is obtained based on the allocation cost.

[0021] In one embodiment, updating the allocation cost according to the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the multiple target edge computing nodes, and the allocation relationship value includes:

[0022] Get the number of iterations;

[0023] Calculating a partial derivative of the allocation cost with respect to the allocation relationship value;

[0024] updating the distribution relationship value according to the number of iterations and the partial derivative;

[0025] The allocation cost is calculated according to the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the multiple target edge computing nodes, and the updated allocation relationship value.

[0026] In one embodiment, determining the execution time corresponding to the remaining garment processing tasks includes:

[0027] If the task type of the remaining clothing processing task is a cutting type, obtaining an operation image corresponding to the remaining clothing processing task;

[0028] Based on the processing frequency of the worker's operation in the operation image, the execution time corresponding to the remaining garment processing tasks is determined.

[0029] In one embodiment, obtaining the current delay time of the target edge computing node whose processing time exceeds the preset time includes:

[0030] Determine the clothing category of the clothing processing task corresponding to the target edge computing node whose processing time exceeds the preset time;

[0031] Identify the historical clothing category that is the same as the clothing category in question;

[0032] Obtaining a historical costume processing task corresponding to the historical costume category;

[0033] Determine a target historical clothing processing task in the historical clothing processing tasks that has the same task type as the clothing processing task corresponding to the target edge computing node whose processing time exceeds the preset time;

[0034] Calculating the average processing time corresponding to a plurality of target historical clothing processing tasks; wherein the average processing time is the total required time for clothing processing tasks of the same task type as the target historical clothing processing task;

[0035] Based on the multiple average processing times, a current delay time of the target edge computing node whose processing time exceeds the preset time is determined.

[0036] In one embodiment, determining, based on the multiple average processing times, the current delay time of the target edge computing node whose processing time exceeds the preset time includes:

[0037] Calculating the sum of the multiple average processing times as the total required time of the target edge computing node whose processing time exceeds the preset time;

[0038] Obtaining the total execution time of the target edge computing node whose processing time exceeds the preset time;

[0039] Based on the total demand time of the target edge computing node whose processing time exceeds the preset time and the total execution time of the target edge computing node whose processing time exceeds the preset time, the current delay time of the target edge computing node whose processing time exceeds the preset time is determined.

[0040] In one embodiment, after determining the optimal target edge computing node for allocating the remaining garment processing tasks based on the allocation relationship value, the scheduling method for the garment production line further includes:

[0041] Get the execution time of the garment processing task corresponding to each garment order, which is of process type;

[0042] Get the execution time of the tailoring type clothing processing task corresponding to each clothing order;

[0043] The ratio of the execution time of the craft type to the execution time of the tailoring type corresponding to each clothing order is displayed in the form of a progress bar.

[0044] According to another aspect of the present invention, there is provided a scheduling device for a garment production line, comprising:

[0045] Acquisition module, used to obtain multiple clothing orders;

[0046] An allocation module is used to allocate the same clothing processing tasks in each clothing order to the corresponding target edge computing node; wherein the target edge computing node is used to process the same clothing processing tasks in sequence;

[0047] The scheduling module is configured to obtain the remaining garment processing tasks corresponding to the target edge computing node if the processing time of the target edge computing node exceeds a preset time; wherein the remaining garment processing tasks are tasks that the target edge computing node has not completed within the processing time; obtain the allocation relationship value between the remaining garment processing tasks and other target edge computing nodes; and determine the optimal target edge computing node to which the remaining garment processing tasks are allocated based on the allocation relationship value.

[0048] The beneficial effects of the present invention are reflected in that the present invention provides a scheduling method and device for a clothing production line, including: obtaining multiple clothing orders, allocating the same clothing processing tasks in each clothing order to the corresponding target edge computing node, wherein the target edge computing node is used to process the same clothing processing tasks in sequence, and if the processing time of the target edge computing node exceeds the preset time, obtaining the remaining clothing processing tasks corresponding to the target edge computing node; wherein the remaining clothing processing tasks are tasks that the target edge computing node has not completed within the processing time, obtaining the allocation relationship value between the remaining clothing processing tasks and other target edge computing nodes, and determining the optimal target edge computing node for the allocation of the remaining clothing processing tasks based on the allocation relationship value. By detecting the computing delay of the edge computing node and based on the allocation relationship value, the computing power of all edge computing nodes in the production system is automatically rescheduled, thereby avoiding the problems of idle computing power and excessive load on individual devices in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flowchart of a scheduling method for a clothing production line provided by an exemplary embodiment of the present invention.

[0050] Figure 2 It is a structural diagram of task allocation provided by an exemplary embodiment of the present invention.

[0051] Figure 3 It is a flowchart of determining the optimal target edge computing node provided by an exemplary embodiment of the present invention.

[0052] Figure 4 It is a structural diagram of a scheduling device for a clothing production line provided by an exemplary embodiment of the present invention.

[0053] Figure 5 It is a structural schematic diagram of a scheduling device for a clothing production line provided by another exemplary embodiment of the present invention.

[0054] Figure 6 is a structural diagram of an electronic device provided by an exemplary embodiment of the present invention.

[0055] Figure numerals: 201, acquisition module; 202, allocation module; 203, scheduling module; 2031, first determination unit; 2032, time acquisition unit; 2033, second determination unit; 10, electronic device; 11, processor; 12, memory; 13, input device; 14, output device. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0057] Example 1: Figure 1 It is a flowchart of a scheduling method for a clothing production line provided by an exemplary embodiment of the present invention. Figure 2 FIG. 1 is a schematic diagram of a structure of a task allocation provided by an exemplary embodiment of the present invention. Figure 1-2 As shown in Figure 2, the scheduling method for the clothing production line includes:

[0058] S110: Acquire multiple clothing orders.

[0059] S120: Allocate the same clothing processing tasks in each clothing order to the corresponding target edge computing node, wherein the target edge computing node is used to process the same clothing processing tasks in sequence.

[0060] In an embodiment of the present invention, each clothing order contains clothing from different clothing categories, and clothing from different clothing categories requires different clothing processing tasks. For example, clothing categories may include pants, skirts, short sleeves, etc. Clothing processing tasks include design tasks, process tasks, and cutting tasks. Process tasks include ironing and rinsing, and cutting tasks include cutting, machine sewing, and embroidery. The same clothing processing tasks are identified from multiple clothing orders and assigned to the corresponding target edge computing nodes. The target edge computing nodes process the clothing processing tasks uploaded in sequence.

[0061] like Figure 2 As shown, the purple circle represents the embroidery process task, the red circle represents the washing process task, the square corresponding to the purple circle represents the edge computing node responsible for analyzing the embroidery process task, and the square corresponding to the red circle represents the edge computing node responsible for the washing process task.

[0062] S130: If the processing time of the target edge computing node exceeds the preset time, the remaining clothing processing tasks corresponding to the target edge computing node are obtained, wherein the remaining clothing processing tasks are tasks that are not completed by the target edge computing node within the processing time.

[0063] In an embodiment of the present invention, if the processing time of the target edge computing node exceeds the preset time, it means that the target edge computing node has been delayed, and the remaining clothing processing tasks in the target edge computing node need to be scheduled to prevent other edge computing nodes from being idle and the target edge computing node from being congested in task processing.

[0064] S140: Obtain the distribution relationship value between the remaining clothing processing tasks and other target edge computing nodes.

[0065] S150: Determine the optimal target edge computing node for allocating the remaining clothing processing tasks based on the allocation relationship value.

[0066] Example 2: In one embodiment, S140 can be specifically implemented as follows: determining the execution time corresponding to the remaining clothing processing tasks; determining the computing power value of the remaining clothing processing tasks allocated to other target edge computing nodes; and determining the allocation relationship value between the remaining clothing processing tasks and the other target edge computing nodes based on the execution time corresponding to the remaining clothing processing tasks and the computing power value of the remaining clothing processing tasks allocated to other target edge computing nodes.

[0067] In an embodiment of the present invention, determining the execution time corresponding to the remaining clothing processing task may include the following steps: if the task type of the remaining clothing processing task is a cutting type, obtaining an operation image corresponding to the remaining clothing processing task; and determining the execution time corresponding to the remaining clothing processing task based on the processing frequency of the worker's operation in the operation image.

[0068] In the present invention, garment processing tasks differ in their execution times due to the different processing methods. If the remaining garment processing tasks are of the cutting type, the corresponding operation images are acquired. These images are captured by a camera while workers are performing the processing. By analyzing each worker's processing frequency, the execution times of the remaining garment processing tasks can be determined. It should be understood that cutting tasks primarily rely on worker operation, and the operation model is relatively fixed. They primarily use specific tools to process fabric or garments, but the processing location varies. For example, the specific tool is a pair of scissors.

[0069] Specifically, the operator's tool is determined through the operation image and its type is identified. For example, the tool could be a pair of scissors or a needle. Based on the tool type, the worker's processing frequency is determined. For example, if the tool is a pair of scissors, the number of times the scissors are closed and opened on the garment is considered a processing frequency. Once the worker completes the cutting, the number of times the scissors are closed and opened is counted as the worker's processing frequency during cutting.

[0070] If the remaining garment processing tasks are of the process type, the corresponding operation time of the process type task can be directly obtained as its execution time. Specifically, the process data unit is responsible for collecting process processing progress data, such as ironing and rinsing. These processes usually have clear operation times, so the execution time of the target process can be determined by calculating its operation time.

[0071] If the task type of the remaining garment processing task is a design type, the operation time uploaded by the designer or pattern maker is the execution time of the remaining garment processing task.

[0072] After obtaining the execution time of the remaining garment processing tasks, the computing capacity values ​​of other target edge computing nodes for allocating the remaining garment processing tasks are obtained. Computing capacity refers to the ability of each edge computing node to execute tasks per unit time. When the remaining garment processing tasks cannot be processed on the original node, these tasks need to be transferred to other target edge computing nodes. To allocate tasks appropriately, the computing capacity values ​​of the target nodes need to be evaluated. A node's computing capacity determines the speed, efficiency, or potential latency of its task processing, which in turn influences task allocation decisions. For example, there are multiple target edge computing nodes, each with different computing capacities. Specifically, node A has a computing capacity value of 15 (meaning it can process 15 tasks per second). Node B has a computing capacity value of 10 (meaning it can process 10 tasks per second). Node C has a computing capacity value of 5 (meaning it can process 5 tasks per second). If the remaining tasks need to be processed within the unit time and the current node cannot handle all tasks, tasks can be allocated based on the computing capacity values ​​of these target nodes. For example, more tasks can be allocated to the node with greater computing capacity (node ​​A), followed by nodes B and C.

[0073] Example 3: Figure 3 This is a flow chart of determining the optimal target edge computing node provided by an exemplary embodiment of the present invention. Figure 3 , S150 can include:

[0074] S151: Obtain the current delay time of the target edge computing node whose processing time exceeds the preset time.

[0075] S152: Obtain the average delay time of multiple target edge computing nodes.

[0076] S153: Determine the optimal target edge computing node for allocating the remaining clothing processing tasks based on the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of multiple target edge computing nodes, and the allocation relationship value.

[0077] In an embodiment of the present invention, the current delay time of the target edge computing node whose processing time exceeds the preset time is updated according to the current delay time of the target edge computing node whose processing time exceeds the preset time and the average delay time of multiple target edge computing nodes.

[0078] Specifically, the calculation formula for the updated current delay time of the target edge computing node is:

[0079] ,in, is the learning rate, k represents the number of iterations, Indicates the current delay time of the target edge computing node whose processing time exceeds the preset time. Represents the average delay time of multiple target edge computing nodes, The plurality of target edge computing nodes represents all target edge computing nodes.

[0080] It is understandable that the delay of the current node i Average latency to all nodes There is a gap .gap Indicates whether the latency of node i is higher or lower than the average. If the difference is positive, it means that the latency of node i is higher than the average; if the difference is negative, it means that the latency of node i is lower than the average. The difference is multiplied by a second weight to obtain the adjustment amount for the latency of node i. If the latency of node i is higher than the average, then is positive, the formula Will reduce , so that it tends to the average value If the delay of node i is lower than the average, then is negative, the formula Will increase , making it tend towards the average value.

[0081] parameter Control the size of the adjustment step, if If the value is large, the adjustment range of each iteration will be large, but it may cause oscillation or instability; if The smaller the value, the smaller the adjustment range of each iteration, and the smoother and more stable the adjustment process. In this way, through successive updates, the delay of each node gradually approaches the average delay, achieving overall load balancing of the system. By executing the formula multiple times , each node The system will continue to adjust, eventually tending towards the average latency of the system. This adjustment process can effectively balance the workload of each node in the system, preventing some nodes from being overloaded while others are idle, thus achieving the goal of load balancing.

[0082] Example 4: In one embodiment, S153 can be specifically implemented as follows: updating the allocation cost based on the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of multiple target edge computing nodes, and the allocation relationship value; if the allocation cost is less than the preset cost threshold, then based on the allocation cost, obtaining the optimal target edge computing node for the remaining clothing processing task allocation.

[0083] In an embodiment of the present invention, the current delay time of other target edge computing nodes after update can be obtained based on the current delay time of the target edge computing node whose processing time exceeds the preset time and the average delay time of multiple target edge computing nodes, and then the allocation cost can be calculated in combination with the allocation relationship value.

[0084] The calculation formula for allocation cost is: ,in, is the first weight, M is 0, 1…, M, M represents the task assigned to the remaining clothing processing The total number of computing capabilities of other target edge computing nodes, N is 0, 1..., N, N represents the total number of target edge computing nodes whose processing time exceeds the preset time. Indicates the current delay time of the target edge computing node whose processing time exceeds the preset time. Represents the average delay time of multiple target edge computing nodes, represents the second weight, Remaining clothing processing tasks The corresponding execution time, Indicates the tasks assigned to the remaining garments The computing power of other target edge computing nodes.

[0085] If the allocation cost value is the minimum, the optimal target edge computing node for the remaining clothing processing tasks is determined by the allocation relationship value in the allocation cost value.

[0086] Example 5: In one embodiment, S153 may be specifically implemented as follows: obtaining the number of iterations; calculating the partial derivative of the allocation cost with respect to the allocation relationship value; updating the allocation relationship value based on the number of iterations and the partial derivative; and calculating the allocation cost based on the current delay of the target edge computing node whose processing time exceeds a preset time, the average delay of multiple target edge computing nodes, and the updated allocation relationship value. By minimizing the objective function C, the present invention ultimately achieves reasonable task allocation and system load balancing.

[0087] In this embodiment of the present invention, the remaining clothing processing task and other target edge computing nodes The distribution relationship uses binary decision variables Indicates. When it is 1, it means there are remaining clothing processing tasks Assign to other target edge computing nodes .when When it is 0, it means that the allocation is not completed.

[0088] Specifically, the calculation formula of the allocation cost is converted into continuous variable optimization through the linear relaxation method, and the calculation formula of the allocation relationship value can be obtained as follows:

[0089] ;in, is the learning rate, Is the allocation cost about The partial derivative of , k represents the number of iterations.

[0090] According to the formula It can be obtained that in each iteration, it is necessary to calculate the objective function C for each distribution coefficient in the current case Gradient . Then multiply by the learning rate , controls the step size of each update. Choose the appropriate That is, it can converge quickly without oscillation. Finally, use the current distribution coefficient Subtract gradient adjustment , get the distribution coefficient for the next iteration step .

[0091] In one embodiment, before S130, the scheduling method of the clothing production line can be specifically implemented as follows: determining the clothing category of the clothing processing task corresponding to the target edge computing node whose processing time exceeds the preset time; determining the historical clothing category that is the same as the clothing category; obtaining the historical clothing processing tasks corresponding to the historical clothing category; wherein the historical clothing processing tasks are the same as the task type of the clothing processing tasks; calculating the average processing time corresponding to multiple historical clothing processing tasks; wherein the average processing time is the total demand time of the clothing processing tasks corresponding to the target edge computing node whose processing time exceeds the preset time; based on multiple average processing times, determining the current delay time of the target edge computing node whose processing time exceeds the preset time.

[0092] In this embodiment of the present invention, upon receiving an order, the order's category (tops / pants / dresses, etc.) and process complexity (embroidery / printing / washing, etc.) are automatically analyzed. The production capacity requirements for the target production volume in the order are then calculated for each production line. This capacity requirement is calculated by taking the average processing time for a single garment using a single process (the average processing time can be obtained from historical data). Thus, for a single order, the total production requirements for each garment category across different processing stages are determined.

[0093] For clothing processing tasks corresponding to target edge computing nodes whose processing time exceeds a preset time, there are different clothing categories. Therefore, it is necessary to determine the clothing category corresponding to the clothing processing tasks, such as pants, dresses, skirts, etc. Then, a historical clothing category with the same clothing category is determined. Historical clothing processing tasks with the same historical clothing category are obtained. From these historical clothing processing tasks, target historical clothing processing tasks with the same task type as clothing processing tasks corresponding to other target edge computing nodes are identified. The average processing time corresponding to these multiple target historical clothing processing tasks is then calculated. For example, if clothing processing task A for a target edge computing node whose processing time exceeds a preset time has a clothing category of pants, a historical clothing processing task B for pants is determined. From these historical clothing processing tasks, target historical clothing processing tasks C with the same task type as clothing processing task A are selected. The task execution times of multiple target historical clothing processing tasks C are obtained, and their average processing time is calculated. This average processing time is the total required time for clothing processing task A. Because the clothing processing tasks corresponding to the target edge computing node whose processing time exceeds the preset time have different clothing categories, there are multiple average processing times. Calculating these multiple average processing times yields the total required time for the target edge node whose processing time exceeds the preset time.

[0094] In one embodiment, the scheduling method of the clothing production line can be specifically implemented as follows: calculating the sum of multiple average processing times as the total demand time of the target edge computing node whose processing time exceeds the preset time; obtaining the total execution time of the target edge computing node whose processing time exceeds the preset time; and determining the current delay time of the target edge computing node whose processing time exceeds the preset time based on the total demand time of the target edge computing node whose processing time exceeds the preset time and the total execution time of the target edge computing node whose processing time exceeds the preset time.

[0095] In this embodiment of the present invention, if the total demand time of the target edge computing node whose processing time exceeds the preset time is less than the total execution time of the target edge computing node whose processing time exceeds the preset time, it indicates that the target edge computing node whose processing time exceeds the preset time has experienced a delay. The difference between the total execution time and the total demand time is the current delay time of the target edge computing node whose processing time exceeds the preset time.

[0096] In one embodiment, S150 can be specifically implemented as follows: obtaining the execution time of the type of clothing processing task corresponding to each clothing order as the craft type; obtaining the execution time of the type of clothing processing task corresponding to each clothing order as the cutting type; and displaying the ratio between the execution time of the craft type and the execution time of the cutting type corresponding to each clothing order in the form of a progress bar.

[0097] In an embodiment of the present invention, the ratio between the execution time of the craft-type clothing processing task corresponding to each clothing order and the execution time of the tailoring-type clothing processing task corresponding to each clothing order is calculated, and then the results are displayed in the form of a progress bar according to each order. For each clothing order, the design-type clothing processing task is displayed according to the value reported by the number of personnel.

[0098] In this embodiment of the present invention, relevant staff upload and visualize the progress of their design or pattern making. The uploaded data includes the order number and progress (as a percentage). For process-related processing tasks, the present invention uses a barcode scanner to scan the production barcode of the garment or fabric being processed to obtain identification information. When the process-related equipment starts up, the equipment's operating data and identity information are uploaded to the associated edge computing device. Equipment operating data includes: equipment startup time and planned equipment operating hours. For cutting processing tasks, a camera captures image data and uploads it to the edge computing device.

[0099] Figure 4 Schematic diagram of the structure of a scheduling device for a clothing production line provided by an exemplary embodiment of the present invention. Figure 4 The scheduling device of the clothing production line includes: an acquisition module 201 for acquiring multiple clothing orders; an allocation module 202 for allocating the same clothing processing tasks in each clothing order to the corresponding target edge computing node; wherein the target edge computing node is used to process the same clothing processing tasks in sequence; a scheduling module 203 for acquiring the remaining clothing processing tasks corresponding to the target edge computing node if the processing time of the target edge computing node exceeds the preset time; wherein the remaining clothing processing tasks are the tasks that the target edge computing node has not completed within the processing time; acquiring the allocation relationship value between the remaining clothing processing tasks and other target edge computing nodes; and determining the optimal target edge computing node for allocating the remaining clothing processing tasks based on the allocation relationship value.

[0100] Figure 5 1 is a schematic diagram of a scheduling device for a clothing production line according to another exemplary embodiment of the present invention. Figure 5 The scheduling module 203 may include: a first determining unit 2031, used to determine the execution time corresponding to the remaining clothing processing tasks; determine the computing power value of the remaining clothing processing tasks allocated to other target edge computing nodes; and determine the allocation relationship value between the remaining clothing processing tasks and the other target edge computing nodes based on the execution time corresponding to the remaining clothing processing tasks and the computing power value of the remaining clothing processing tasks allocated to other target edge computing nodes.

[0101] In one embodiment, the scheduling module 203 may include: a time acquisition unit 2032, used to obtain the current delay time of other target edge computing nodes; obtain the average delay time of multiple target edge computing nodes; a second determination unit 2033, used to determine the optimal target edge computing node for the remaining clothing processing tasks based on the current delay time of other target edge computing nodes, the average delay time of multiple target edge computing nodes and the allocation relationship value.

[0102] In one embodiment, the second determination unit 2033 can be specifically configured as follows: updating the allocation cost based on the current delay time of other target edge computing nodes, the average delay time of multiple target edge computing nodes, and the allocation relationship value; if the allocation cost is less than the preset cost threshold, then based on the allocation cost, obtaining the optimal target edge computing node for the remaining clothing processing task allocation.

[0103] In one embodiment, the second determination unit 2033 can be specifically configured as: obtaining the number of iterations; calculating the partial derivative of the allocation cost with respect to the allocation relationship value; updating the allocation relationship value based on the number of iterations and the partial derivative; calculating the allocation cost based on the current delay time of other target edge computing nodes, the average delay time of multiple target edge computing nodes, and the updated allocation relationship value.

[0104] In one embodiment, the first determination unit 2031 can be specifically configured as follows: if the task type of the remaining clothing processing task is a cutting type, then obtaining the operation image corresponding to the remaining clothing processing task; based on the processing frequency of the worker's operation in the operation image, determining the execution time corresponding to the remaining clothing processing task.

[0105] In one embodiment, the time acquisition unit 2032 can be specifically configured to: determine the clothing category of the clothing processing tasks corresponding to other target edge computing nodes; determine the historical clothing category that is the same as the clothing category; obtain the historical clothing processing tasks corresponding to the historical clothing category; determine the target historical clothing processing tasks that have the same task type as the clothing processing tasks corresponding to other target edge computing nodes in the historical clothing processing tasks; calculate the average processing time corresponding to multiple target historical clothing processing tasks; wherein the average processing time is the total required time of the clothing processing tasks with the same task type as the target historical clothing processing tasks; based on multiple average processing times, determine the current delay time of other target edge computing nodes

[0106] In one embodiment, the time acquisition unit 2032 can be specifically configured to: calculate the sum of multiple average processing times as the total required time of other target edge computing nodes; obtain the total execution time of other target edge computing nodes; and determine the current delay time of other target edge computing nodes based on the total required time of other target edge computing nodes and the total execution time of other target edge computing nodes.

[0107] In one embodiment, after the scheduling module 203, the scheduling device of the clothing production line can be specifically configured as follows: obtaining the execution time of the type of clothing processing task corresponding to each clothing order as the process type; obtaining the execution time of the type of clothing processing task corresponding to each clothing order as the cutting type; and displaying the ratio of the execution time of the process type to the execution time of the cutting type corresponding to each clothing order in the form of a progress bar.

[0108] Figure 6 The figure shows a block diagram of an electronic device according to an embodiment of the present application.

[0109] like Figure 6 As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0110] The processor 11 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0111] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may execute the program instructions to implement the above-described scheduling methods for clothing production lines of the various embodiments of the present application and / or other desired functions. The computer-readable storage medium may also store various contents such as input signals, signal components, and noise components.

[0112] In one example, the electronic device 10 may further include an input device 13 and an output device 14 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0113] When the electronic device 10 is a stand-alone device, the input device 13 may be a communication network connector, configured to receive collected input signals from the first device and the second device.

[0114] In addition, the input device 13 may also include, for example, a keyboard, a mouse, and the like.

[0115] The output device 14 can output various information to the outside, including determined distance information, direction information, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0116] Of course, to simplify, Figure 6 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device 10 may further include any other appropriate components according to specific application scenarios.

[0117] The computer program product may be written in any combination of one or more programming languages ​​to implement the program code for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0118] Computer-readable storage media can be any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0119] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A scheduling method for a clothing production line, characterized in that: include: Get multiple clothing orders; Assigning identical garment processing tasks in each garment order to corresponding target edge computing nodes; wherein the target edge computing nodes are used to sequentially process identical garment processing tasks; If the processing time of the target edge computing node exceeds the preset time, the remaining clothing processing tasks corresponding to the target edge computing node are obtained; wherein the remaining clothing processing tasks are tasks that the target edge computing node has not completed within the processing time; Determining the execution time corresponding to the remaining clothing processing tasks; Determine the computing capacity values ​​of the remaining clothing processing tasks allocated to other target edge computing nodes; Determine the allocation relationship value between the remaining clothing processing tasks and other target edge computing nodes according to the execution time corresponding to the remaining clothing processing tasks and the computing capacity value of the remaining clothing processing tasks allocated to other target edge computing nodes; Get the current delay time of the target edge computing node whose processing time exceeds the preset time; Get the average latency of multiple target edge computing nodes; Get the number of iterations; Calculating a partial derivative of the allocation cost with respect to the allocation relationship value; updating the distribution relationship value according to the number of iterations and the partial derivative; Calculating an allocation cost according to a current delay time of the target edge computing node whose processing time exceeds a preset time, an average delay time of the plurality of target edge computing nodes, and an updated allocation relationship value; If the allocation cost is less than a preset cost threshold, the optimal target edge computing node for the allocation of the remaining clothing processing tasks is obtained based on the allocation cost.

2. The scheduling method for a clothing production line according to claim 1, characterized in that: Determining the execution time corresponding to the remaining clothing processing tasks includes: If the task type of the remaining clothing processing task is a cutting type, obtaining an operation image corresponding to the remaining clothing processing task; Based on the processing frequency of the worker's operation in the operation image, the execution time corresponding to the remaining garment processing tasks is determined.

3. The scheduling method for a clothing production line according to claim 1, characterized in that: The obtaining of the current delay time of the target edge computing node whose processing time exceeds the preset time includes: Determine the clothing category of the clothing processing task corresponding to the target edge computing node whose processing time exceeds the preset time; Identify the historical clothing category that is the same as the clothing category in question; Obtaining a historical costume processing task corresponding to the historical costume category; Determine a target historical clothing processing task in the historical clothing processing tasks that has the same task type as the clothing processing task corresponding to the target edge computing node whose processing time exceeds the preset time; Calculating the average processing time corresponding to a plurality of target historical clothing processing tasks; wherein the average processing time is the total required time for clothing processing tasks of the same task type as the target historical clothing processing task; Based on the multiple average processing times, a current delay time of the target edge computing node whose processing time exceeds the preset time is determined.

4. The scheduling method for a clothing production line according to claim 3, characterized in that: Determining, based on the multiple average processing times, the current delay time of the target edge computing node whose processing time exceeds the preset time includes: Calculating the sum of the multiple average processing times as the total required time of the target edge computing node whose processing time exceeds the preset time; Obtaining the total execution time of the target edge computing node whose processing time exceeds the preset time; Based on the total demand time of the target edge computing node whose processing time exceeds the preset time and the total execution time of the target edge computing node whose processing time exceeds the preset time, the current delay time of the target edge computing node whose processing time exceeds the preset time is determined.

5. The scheduling method for a clothing production line according to claim 1, characterized in that: After determining the optimal target edge computing node for allocating the remaining clothing processing tasks based on the allocation relationship value, the method further includes: Get the execution time of the garment processing task corresponding to each garment order, which is of process type; Get the execution time of the tailoring type clothing processing task corresponding to each clothing order; The ratio between the execution time of the craft type and the execution time of the tailoring type corresponding to each clothing order is displayed in the form of a progress bar.

6. A scheduling device for a clothing production line, characterized in that: include: Acquisition module, used to obtain multiple clothing orders; An allocation module is used to allocate the same clothing processing tasks in each clothing order to the corresponding target edge computing node; wherein the target edge computing node is used to process the same clothing processing tasks in sequence; a scheduling module configured to, if the processing time of the target edge computing node exceeds a preset time, obtain remaining garment processing tasks corresponding to the target edge computing node; wherein the remaining garment processing tasks are tasks that the target edge computing node has not completed within the processing time; determine the execution time corresponding to the remaining garment processing tasks; determine the computing power values ​​of the remaining garment processing tasks allocated to other target edge computing nodes; determine the allocation relationship values ​​between the remaining garment processing tasks and other target edge computing nodes based on the execution time corresponding to the remaining garment processing tasks and the computing power values ​​of the remaining garment processing tasks allocated to other target edge computing nodes; obtain a current delay time of the target edge computing node whose processing time exceeds the preset time; obtain an average delay time of multiple target edge computing nodes; obtain the number of iterations; calculate a partial derivative of an allocation cost with respect to the allocation relationship value; update the allocation relationship value based on the number of iterations and the partial derivative; calculate an allocation cost based on the current delay time of the target edge computing node whose processing time exceeds the preset time, the average delay time of the multiple target edge computing nodes, and the updated allocation relationship value; and if the allocation cost is less than a preset cost threshold, obtain an optimal target edge computing node for allocation of the remaining garment processing tasks based on the allocation cost.

Images