Production scheduling optimization method and system in packaging and printing factory

Abstract

The invention relates to a production scheduling optimization method and system in a packaging and printing factory. The method comprises the steps of obtaining time constraints of tasks in a plurality of orders to be scheduled and resource consumption of the tasks; based on a resource inventory and the resource consumption, adjusting a default schedule to obtain a first schedule; and determining a target scheduling algorithm by using an order scheduling model based on the resource consumption, the first scheduling and the time constraint, and sorting the plurality of orders based on the target scheduling algorithm to obtain a scheduling result. According to the invention, in the scheduling process, the resource inventory, the customer demand (order) and the production capacity are considered, the variable relationship among production, supply and demand can be coped, and the time constraint of each task in the order is also considered, so that the probability of obtaining the optimal order scheduling which reasonably utilizes resources, saves time and meets the customer demand is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a production scheduling optimization method and system in a packaging and printing factory. Background Art

[0002] In the manufacturing industry, it is very important to schedule orders reasonably and produce according to the schedule. Reasonable scheduling can help companies improve production efficiency, reduce production costs, promote multi-departmental collaboration within the company, and improve customer satisfaction.

[0003] Order scheduling involves effective resource allocation, reasonable production plan formulation, and timely satisfaction of customer needs, and is a crucial link in the manufacturing industry. However, there are many factors to consider in the scheduling process; the relationship between production, supply, and demand is flexible and changeable, making it difficult to obtain reasonable and accurate scheduling results in a timely manner. Summary of the invention

[0004] In view of this, the present disclosure proposes a packaging printing order scheduling solution.

[0005] According to one aspect of the present disclosure, a packaging printing order scheduling method is provided, including: obtaining the time constraints of each task in a plurality of orders to be scheduled and the resource consumption of the tasks; adjusting the default schedule based on the resource inventory and the resource consumption to obtain a first schedule; determining a target scheduling algorithm using an order scheduling model based on the resource consumption, the first schedule and the time constraints, and sorting the plurality of orders based on the target scheduling algorithm to obtain a scheduling result.

[0006] In a possible implementation, the method of obtaining the time constraints for each task in the multiple orders to be scheduled includes: obtaining the personnel time, material time, and equipment time occupied by each of the tasks; for a single task, determining the maximum value of the material time and the equipment time; taking the sum of the maximum value and the personnel time of the single task as the execution time of the single task, so as to obtain the execution time of each of the tasks; obtaining the time difference between the start time of each of the tasks and the start time of their previous tasks, and taking the time difference as the lag time of each of the tasks.

[0007] In a possible implementation, the default schedule is adjusted based on the resource inventory and the resource consumption to obtain the first schedule, including: traversing the resource consumption of the multiple orders according to the default schedule, accumulating the resource consumption of each order and determining the resource inventory balance; when the resource inventory balance is insufficient, in response to the received inventory replenishment, a new resource inventory balance is obtained, and the traversal is continued until the resource consumption of all orders is traversed; when the resource inventory balance is insufficient, in response to the received skip order operation, the current order is filtered out, and the traversal is continued until the resource consumption of all orders is traversed; during the traversal, if only the inventory replenishment is received, the default schedule is used as the first schedule; during the traversal, if at least one skip order operation is received, the filtered orders are removed from the default schedule to obtain the first schedule.

[0008] In one possible implementation, the training process of the order scheduling model includes: obtaining an initial model, in which multiple scheduling algorithms are preset; using the initial model to determine a preferred algorithm based on a single training sample; using the preferred algorithm to schedule orders in the single training sample to obtain a first scheduling result; determining the similarity between the first scheduling result and the true value of the single training sample; determining a reward value based on the similarity; adjusting the probability values ​​corresponding to the multiple scheduling algorithms based on the reward value until the similarity is not less than a similarity threshold or the multiple scheduling algorithms are all determined to be preferred algorithms at least once, and using the initial model to process other training samples; until the model convergence condition is met, the order scheduling model is obtained.

[0009] In a possible implementation, the method further includes: determining the sample resource consumption of each sample task based on the sample personnel time, sample material time, and sample equipment time occupied by each sample task of multiple sample orders, as well as the unit time personnel consumption, unit time material consumption, and unit time equipment consumption of each sample task; determining the sample execution time of each sample task based on the sample personnel time, sample material time, and sample equipment time; taking the sample time difference between the start time of the sample task and the start time of the previous sample task as the sample lag time of the sample task, and taking the sample execution time and the sample lag time as the sample time constraints of the sample task; taking the order of accepting each sample task as the sample schedule, and taking the sample resource consumption, sample time constraints, and sample schedule corresponding to the multiple sample orders as a single training sample.

[0010] In a possible implementation, the method further includes: reading a plurality of initial orders; detecting the initial orders and removing empty orders to obtain the plurality of orders to be scheduled.

[0011] In a possible implementation, the method further includes: sorting the first orders executed in parallel in the scheduling result according to the order of receiving the orders, and updating the scheduling result to obtain an updated scheduling result.

[0012] According to another aspect of the present disclosure, a packaging printing order scheduling system is provided, comprising:

[0013] A time constraint and resource consumption determination unit, used to obtain the time constraint of each task in a plurality of orders to be scheduled and the resource consumption of each task;

[0014] A first schedule determining unit, configured to adjust a default schedule based on resource inventory and resource consumption to obtain a first schedule;

[0015] The scheduling result determination unit is used to determine a target scheduling algorithm based on the resource consumption, the first schedule and the time constraint using an order scheduling model, and to sort the multiple orders based on the target scheduling algorithm to obtain a scheduling result.

[0016] In a possible implementation manner, the time constraint and resource consumption determination unit is further configured to:

[0017] Obtaining the personnel time, material time, and equipment time occupied by each of the tasks;

[0018] For a single task, determine the maximum value of material duration and equipment duration;

[0019] The sum of the maximum value and the personnel time of the single task is taken as the execution time of the single task to obtain the execution time of each task;

[0020] The time difference between the start time of each task and the start time of the previous task is obtained, and the time difference is used as the lag time of each task.

[0021] In a possible implementation manner, the first schedule determination unit is further configured to:

[0022] According to the default schedule, traverse the resource consumption of the multiple orders, accumulate the resource consumption of each order and determine the resource inventory balance;

[0023] In the case where the resource inventory balance is insufficient, a new resource inventory balance is obtained in response to the received inventory replenishment amount, and the traversal is continued until the resource consumption of all orders is traversed;

[0024] In the case where the resource inventory surplus is insufficient, in response to the received skip order operation, the current order is screened out, and the traversal is continued until the resource consumption of all orders is traversed;

[0025] In the traversal, when only the inventory replenishment quantity is received, the default schedule is used as the first schedule;

[0026] In the traversal, when at least one skip order operation is received, the filtered orders are removed from the default schedule to obtain the first schedule.

[0027] In a possible implementation, the training process of the order scheduling model includes:

[0028] Acquire an initial model, wherein a plurality of scheduling algorithms are preset in the initial model;

[0029] Using the initial model, determining a preferred algorithm based on a single training sample;

[0030] Using an optimal algorithm to schedule the orders in the single training sample to obtain a first scheduling result;

[0031] Determining a similarity between the first scheduling result and a true value of the single training sample;

[0032] Determining a reward value according to the similarity;

[0033] According to the reward value, the probability values ​​corresponding to the multiple scheduling algorithms are adjusted until the similarity is not less than the similarity threshold or the multiple scheduling algorithms are all determined to be the preferred algorithm at least once, and the initial model is used to process other training samples; until the model convergence condition is met, the order scheduling model is obtained.

[0034] In a possible implementation, the training process of the order scheduling model further includes:

[0035] Determine the sample resource consumption of each sample task based on the sample personnel time, sample material time, and sample equipment time occupied by each sample task of the plurality of sample orders, as well as the personnel consumption per unit time, material consumption per unit time, and equipment consumption per unit time of each sample task;

[0036] Determining the sample execution time of each of the sample tasks based on each sample personnel time, sample material time, and sample equipment time;

[0037] The sample time difference between the start time of the sample task and the start time of the previous sample task is used as the sample lag time of the sample task, and the sample execution time and the sample lag time are used as the sample time constraint of the sample task;

[0038] The order of accepting orders for the sample tasks is used as a sample schedule, and the sample resource consumption, sample time constraints, and sample schedules corresponding to the multiple sample orders are used as a single training sample.

[0039] In a possible implementation, the system further includes:

[0040] An initial order reading unit, used for reading a plurality of initial orders;

[0041] The to-be-scheduled order determination unit is used to detect the initial order and remove empty orders to obtain the multiple orders to be scheduled.

[0042] In a possible implementation, the system further includes:

[0043] The scheduling result updating unit is used to sort the first orders executed in parallel in the scheduling result according to the order of receiving the orders, and update the scheduling result to obtain the updated scheduling result.

[0044] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0045] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions implement the above method when executed by a processor.

[0046] According to another aspect of the present disclosure, a computer program product is provided, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0047] In the disclosed embodiment, in the process of scheduling the orders to be scheduled, the resource inventory (personnel, materials, and equipment), customer demand (orders), and production capacity (equipment and personnel) are taken into consideration, and the changing relationship between production, supply, and demand can be dealt with. In addition, the time constraints of each task in the order are also taken into consideration, so the probability of obtaining the optimal order schedule that reasonably utilizes resources, saves time, and meets customer needs is increased. In addition, the method uses the order scheduling model to directly process the resource consumption, the first schedule, and the time constraints, which can reduce manual participation and improve efficiency.

[0048] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0050] Figure 1 A flowchart of a packaging printing order scheduling method provided in an embodiment of the present disclosure.

[0051] Figure 2 A flowchart of another packaging printing order scheduling method provided in an embodiment of the present disclosure.

[0052] Figure 3 A schematic diagram of the structure of a packaging printing order scheduling system provided in an embodiment of the present disclosure.

[0053] Figure 4 A schematic diagram of the structure of an electronic device for packaging printing order scheduling provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0054] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0055] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0056] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0057] In the manufacturing industry, the contract management department, supply chain department, and production department are interrelated and independent departments. This means that the company needs these three departments or even more departments to work together to obtain and complete orders. Moreover, customers have different requirements for orders, and the resources used for each order may be different in some parts and the same in others. Therefore, it is necessary to schedule orders so that multiple orders can be executed in sequence and achieve the most reasonable state of time and resource consumption.

[0058] Compared with other manufacturing industries, the packaging and printing industry often has the characteristics of multiple varieties and small batches. This characteristic requires order scheduling that can provide high flexibility and responsiveness. Due to fierce market competition, customers have higher and higher requirements for delivery time, and often hope to achieve on-demand manufacturing and on-time delivery. This requires the order scheduling of packaging and printing companies to respond quickly to market changes and ensure that orders are delivered on time.

[0059] In view of the characteristics of the packaging and printing industry, higher requirements are placed on order scheduling in terms of accuracy, efficiency, flexibility and rationality.

[0060] Figure 1 The flowchart of the packaging printing order scheduling method provided by the embodiment of the present disclosure is as follows. Figure 1 As shown, the method includes:

[0061] S11, obtaining the time constraints of each task in a plurality of orders to be scheduled and the resource consumption of each task.

[0062] In the disclosed embodiment, a task may be an item or a step in the process of executing an order. For example, packaging printing may include tasks such as inkjet printing, sheet cutting, and hot stamping. A time constraint may be a restriction on a task in the time dimension. For example, a time constraint may be a restriction on the start time, a restriction on the end time, or a restriction on the time required to complete a task. Another example is that a time constraint may be a restriction on the execution order of two tasks. Resources may include personnel, equipment, materials, etc. In the disclosed embodiment, a single task may be performed by one or more personnel using one or more equipment and consuming one or more materials.

[0063] In the embodiment of the present disclosure, the resource consumption can be represented in the form of a resource requirement matrix. The columns of the resource requirement matrix represent a single resource, and the rows represent a single task. The element in the i-th row and j-th column of the resource requirement matrix represents the consumption of the j-th resource consumed by executing the i-th task.

[0064] The resource requirement matrix can be obtained directly. Alternatively, the time information matrix and the unit resource requirement matrix can be obtained first. Then, the elements of the time information matrix and the unit resource requirement matrix with the same rows and columns are multiplied to obtain the resource requirement matrix.

[0065] Among them, the columns of the time information matrix represent a single resource, and the rows represent a single task. The elements in the sth row and tth column of the time matrix represent the duration of the tth resource occupied by the sth task. The columns of the unit resource demand matrix represent a single resource, and the rows represent a single task. The elements in the sth row and tth column of the unit resource demand matrix represent the consumption of the tth resource consumed by the sth task in unit time.

[0066] S12, adjusting the default schedule based on the resource inventory and the resource consumption to obtain a first schedule.

[0067] The default schedule can be a pre-defined order for executing orders. The default schedule may not be optimal in terms of material consumption and time consumption. For example, the default schedule can be the order in which orders are received.

[0068] Resource inventory can be an estimate of the remaining amount of personnel, equipment, and materials when the to-be-scheduled order begins to be executed, that is, the remaining amount of resources. Resource inventory can be obtained by subtracting the amount of personnel, equipment, and materials in use when the to-be-scheduled order begins to be executed (on non-to-be-scheduled orders) from the total amount of personnel, equipment, and materials.

[0069] In the disclosed embodiment, the resource consumption can be used to update the resource inventory, and the updated resource inventory can be used to adjust the default schedule. For example, orders with low resource consumption can be executed first, and the execution order of orders with low resource consumption can be moved forward. This can not only meet the delivery time requirements of orders with low resource consumption, but also buy time for replenishing resource inventory.

[0070] The default schedule can be adjusted to obtain a first schedule. The orders and order sequence in the first schedule and the default schedule can be the same or different. In the process of obtaining the first schedule, the resource inventory can be replenished. After obtaining the first schedule, the resource inventory can support the execution of the orders in the first schedule.

[0071] S13, based on the resource consumption, the first schedule and the time constraint, determine a target scheduling algorithm using an order scheduling model, and sort the multiple orders based on the target scheduling algorithm to obtain a scheduling result.

[0072] A variety of scheduling algorithms can be pre-stored in the order scheduling model. In the disclosed embodiment, a target scheduling algorithm can be determined based on resource consumption and time constraints. The scheduling method using the target scheduling algorithm is used to schedule the orders to be scheduled, and the obtained scheduling results can obtain the optimal scheduling results in terms of resource consumption, time consumption, and customer demand dimensions.

[0073] The order scheduling model can use a target algorithm to update the first schedule based on the resource consumption, the first schedule and the time constraint to obtain a scheduling result. The scheduling result can include the execution order of the orders to be scheduled and a Gantt chart.

[0074] In the disclosed embodiment, in the process of scheduling the orders to be scheduled, the resource inventory (personnel, materials, and equipment), customer demand (orders), and production capacity (equipment and personnel) are taken into consideration, and the changing relationship between production, supply, and demand can be dealt with. In addition, the time constraints of each task in the order are also taken into consideration, so the probability of obtaining the optimal order schedule that reasonably utilizes resources, saves time, and meets customer needs is increased. In addition, the method uses the order scheduling model to directly process the resource consumption, the first schedule, and the time constraints, which can reduce manual participation and improve efficiency.

[0075] In a possible implementation, the method of obtaining the time constraints for each task in the multiple orders to be scheduled includes: obtaining the personnel time, material time, and equipment time occupied by each of the tasks; for a single task, determining the maximum value of the material time and the equipment time; taking the sum of the maximum value and the personnel time of the single task as the execution time of the single task, so as to obtain the execution time of each of the tasks; obtaining the time difference between the start time of each of the tasks and the start time of their previous tasks, and taking the time difference as the lag time of each of the tasks.

[0076] In the embodiment of the present disclosure, the time constraint includes: the execution time of each task of the order to be scheduled, and the lag time of each task relative to the start time of the previous task.

[0077] For example, a task may include the following multiple stages: personnel operating equipment for pre-preparation, personnel operating equipment for material pre-processing, personnel operating equipment for material processing, and natural placement of materials, and waiting for materials. The above is only an example, and the embodiments of the present disclosure do not limit the specific execution stages of the task.

[0078] In most cases, personnel operate equipment to complete the processing of equipment and materials. This shows that in most cases, personnel and equipment work in parallel, so the personnel time and equipment time are parallel most of the time, and the probability of personnel or equipment operating alone is low. In addition to working in parallel with personnel and equipment, materials also have their own occupied time. Therefore, when determining the execution time of a task, personnel and equipment can be regarded as participating in the task in parallel, and the maximum value of personnel time and equipment time can be determined. On the other hand, the material time can be regarded as a single occupied time. The sum of the maximum value and the material time is taken as the execution time of a single task.

[0079] The time difference can be set manually; the warm-up time of different equipment in the two upstream and downstream tasks can also be used as the time difference. In addition, when the current task depends on the output of the previous task, the current task can only start after the previous task produces a unit of output. Therefore, the time difference can be the time length for the previous task to produce a unit of output.

[0080] In the disclosed embodiment, the maximum value of the personnel time and the material time can be used to calculate the execution time of the task, thereby reducing the probability of repeated calculation of the execution time and improving the accuracy of the execution time. The material time is added to the aforementioned maximum value as a separate occupied time, thereby reducing the probability of underestimation of the execution time and further improving the accuracy of the execution time. In addition, the time constraint includes the execution time of each task and the lag time relative to the previous task, so that the scheduling operation can meet the actual needs of the order in terms of time, thereby improving the rationality and accuracy of the scheduling results.

[0081] In a possible implementation, the default schedule is adjusted based on the resource inventory and the resource consumption to obtain the first schedule, including: traversing the resource consumption of the multiple orders according to the default schedule, accumulating the resource consumption of each order and determining the resource inventory balance; when the resource inventory balance is insufficient, in response to the received inventory replenishment, a new resource inventory balance is obtained, and the traversal is continued until the resource consumption of all orders is traversed; when the resource inventory balance is insufficient, in response to the received skip order operation, the current order is filtered out, and the traversal is continued until the resource consumption of all orders is traversed; during the traversal, if only the inventory replenishment is received, the default schedule is used as the first schedule; during the traversal, if at least one skip order operation is received, the filtered orders are removed from the default schedule to obtain the first schedule.

[0082] In the disclosed embodiment, the resource consumption of each order can be obtained in sequence according to the default scheduling order. Each time the resource consumption of an order is obtained, the resource inventory can be used to subtract the resource consumption of the order to obtain the resource inventory surplus. When the surplus of at least one resource in the resource inventory surplus is less than zero, the user can select an operation. The optional operations include: replenishing the inventory, and skipping the order that makes the resource inventory surplus insufficient.

[0083] If the user chooses to replenish the inventory, the inventory balance is no longer less than zero, and the remaining orders can be traversed until the traversal is completed. If the user chooses to skip the order that makes the resource inventory balance insufficient, the current order is directly filtered out, and the remaining orders are traversed until the traversal is completed. If the inventory balance is insufficient again, the user will continue to make the above selection to complete the traversal.

[0084] When the user chooses to replenish the inventory, the default schedule is not changed, so the default schedule can be used as the first schedule. When the user chooses to skip orders that make the resource inventory balance insufficient, some orders have been screened out, so these screened out orders should be removed from the default schedule to obtain the first schedule.

[0085] In the disclosed embodiment, the resource consumption of each order is traversed based on the inventory balance, so that the resource inventory can meet the demand for each resource of the order to be scheduled. By updating the default schedule to obtain the first schedule, the subsequent scheduling process can be made more effective, the probability of invalid scheduling for orders that cannot participate in the scheduling can be reduced, and the accuracy and availability of the scheduling results can be improved.

[0086] In one possible implementation, the training process of the order scheduling model includes: obtaining an initial model, in which multiple scheduling algorithms are preset; using the initial model to determine a preferred algorithm based on a single training sample; using the preferred algorithm to schedule orders in the single training sample to obtain a first scheduling result; determining the similarity between the first scheduling result and the true value of the single training sample; determining a reward value based on the similarity; adjusting the probability values ​​corresponding to the multiple scheduling algorithms based on the reward value until the similarity is not less than a similarity threshold or the multiple scheduling algorithms are all determined to be preferred algorithms at least once, and using the initial model to process other training samples; until the model convergence condition is met, the order scheduling model is obtained.

[0087] In one example, the preferred algorithm may be an algorithm selected from multiple algorithms preset in the initial model according to a preset algorithm selection order. In another example, multiple algorithms preset in the initial model may be used for scheduling for a single training sample. In the first scheduling, an algorithm may be randomly selected, and in the subsequent scheduling of the single training sample, the selected algorithm may be excluded, and the algorithm may continue to be randomly selected from the remaining algorithms, and the algorithm selected each time may be determined as the preferred algorithm.

[0088] Each time the optimization algorithm is used to schedule the orders in the training sample, a first scheduling result can be obtained. The first scheduling result can be represented by a sequence matrix. The rows of the sequence matrix represent orders and the columns represent the order of execution. The element in the pth row and qth column of the sequence matrix represents that the pth order is scheduled to be executed in the qth order.

[0089] A single training sample may include multiple sample orders and a true value. The most suitable scheduling algorithm is determined manually or by software based on the order situation, time dimension, resource utilization dimension, and customer demand dimension, and the scheduling result determined by the scheduling algorithm is used as the true value of the training sample. Here, the scheduling algorithm determined manually or by software may be one of the multiple scheduling algorithms preset in the initial model. The true value may be expressed in the same form as the first scheduling result. The first scheduling result may be the result of the initial model updating the first schedule based on the training sample.

[0090] Each time a first scheduling result is determined, the first scheduling result can be compared with the true value and the similarity with the true value can be determined. For example, the Euclidean distance or Manhattan distance between the first scheduling result and the true value can be determined. When both the first scheduling result and the true value are represented in matrix form, the ranks of the first scheduling result and the true value can be determined, and the similarity of the two ranks can be determined.

[0091] When the similarity is not less than the similarity threshold, a reward with a value greater than zero is given. When the similarity is less than the similarity threshold, a reward with a value less than zero is given. The similarity threshold can be between 0.91-0.99. In another example, a mapping relationship between a similarity value interval and a reward value can be preset. The value interval of the similarity is determined, and the reward is determined based on the mapping relationship. The similarity and the reward can be positively correlated.

[0092] The reward value is used to adjust the probability value of each scheduling algorithm being used to schedule the sample order. The probability value can be positively correlated with the reward value. If the preset scheduling algorithms are determined as the preferred algorithms at least once and participate in the scheduling, or the similarity is not less than the similarity threshold (that is, the scheduling algorithm used to determine the true value is found), then the training using the current training sample can be terminated. The next training sample is input into the initial model for training until the model convergence condition is met, and the trained model is used as the order scheduling model. Among them, the model convergence condition can be preset. For example: the model convergence condition can be that the training is completed using c training samples, and c is a positive integer. For another example: the model convergence condition can be that the probability of obtaining the same scheduling result as the true value using the first preferred algorithm is greater than the probability threshold.

[0093] In the disclosed embodiment, multiple scheduling algorithms preset in the initial model can be used to schedule sample orders respectively. A first scheduling result can be obtained for each scheduling. The similarity between the first scheduling result and the true value is used to guide the adjustment of the probability values ​​corresponding to each of the multiple scheduling algorithms, so as to improve the probability of selecting a suitable scheduling algorithm according to the resource consumption and time constraints of the order to be scheduled. The training method disclosed in the present invention can enable the trained model to directly obtain the scheduling result according to the resource consumption, time constraints and first scheduling of the order to be scheduled, or update the first scheduling to obtain the scheduling result, thereby improving the efficiency of order scheduling.

[0094] In a possible implementation, it also includes: determining the sample resource consumption of each sample task based on the sample personnel time, sample material time, and sample equipment time occupied by each sample task of multiple sample orders, as well as the unit time personnel consumption, unit time material consumption, and unit time equipment consumption of each sample task; determining the sample execution time of each sample task based on the sample personnel time, sample material time, and sample equipment time; taking the sample time difference between the start time of the sample task and the start time of the previous sample task as the sample lag time of the sample task, and taking the sample execution time and the sample lag time as the sample time constraints of the sample task; taking the order of accepting each sample task as the sample schedule, and taking the sample resource consumption, sample time constraints, and sample schedule corresponding to the multiple sample orders as the single training sample.

[0095] A single training sample may include sample resource consumption, sample time constraints, and sample schedules of multiple sample orders.

[0096] In the embodiment of the present disclosure, for a single training sample, the duration of each sample task occupying resources can be obtained respectively. Then the resource consumption per unit time of each sample task is obtained. As mentioned above, resources may include personnel, materials, and equipment. For the convenience of description, the duration of the sample task occupying personnel is named as sample personnel duration, and the sample personnel duration may include the duration of each personnel occupied by a single sample task; the duration of the sample task occupying materials is named as sample material duration, and the sample material duration may include the duration of each type of material occupied by a single sample task; the duration of the sample task occupying equipment is named as sample equipment duration, and the sample equipment duration may include the duration of each equipment occupied by a single sample task. In addition, the consumption of personnel by the sample task per unit time can be named as personnel consumption per unit time, and the personnel consumption per unit time may include the unit consumption of each personnel required by the sample task; the consumption of materials by the sample task per unit time is named as material consumption per unit time, and the unit material consumption may include the unit consumption of each material required by the sample task; the consumption of equipment by the sample task per unit time is named as equipment consumption per unit time, and the equipment consumption per unit time may include the unit consumption of each equipment required by the sample task.

[0097] In the disclosed embodiment, the duration of a sample task occupying a single resource (sample personnel duration, sample material duration, sample equipment duration) and the consumption of the single resource by the sample task per unit time (personnel consumption per unit time, material consumption per unit time, equipment consumption per unit time) can be multiplied correspondingly to determine the consumption of the single resource by the sample task. Here, a single resource can be a single person, a single device, and a single material. In this way, the sample resource consumption of each sample task in each sample order can be obtained.

[0098] In the disclosed embodiment, the first maximum value of the sample personnel duration and the sample equipment duration corresponding to the sample task can be determined. The sum of the first maximum value and the sample material duration is used as the sample execution duration of the sample task. In addition, the difference between the start time of the sample task and the start time of the previous sample task of the sample task is used as the sample lag duration of the sample task. In addition, the sample execution duration and the sample post-duration of the same sample task are used as the sample time constraint of the sample task.

[0099] In the disclosed embodiment, the sample schedule of the sample task can be manually specified or randomly specified. In this way, the sample resource consumption, sample time constraint, and sample schedule of each sample task in each sample order in multiple sample orders are used as a single training sample. In addition, the optimal scheduling result in the time dimension, resource consumption dimension, and customer demand dimension is used as the true value of the training sample.

[0100] In addition, when a new scheduling algorithm needs to be added to the order scheduling model, the method of constructing training samples in the present disclosure can also be used to train the order scheduling model to obtain an updated order scheduling model.

[0101] By using the method disclosed in the present invention, it is possible to construct adaptive training samples for training an order scheduling model or updating an order scheduling model, thereby improving training efficiency and effect.

[0102] In a possible implementation, the method further includes: reading a plurality of initial orders; detecting the initial orders and removing empty orders to obtain the plurality of orders to be scheduled.

[0103] In the disclosed embodiment, it is possible to detect whether the initial order is an empty order. If it is not an empty order, the initial order is determined as an order to be scheduled, which can reduce the probability of invalid scheduling. In addition, when the shortest duration algorithm is used as the target algorithm, the empty order will be displayed as the first executed order in the scheduling result, which disrupts the scheduling result and reduces the accuracy of the scheduling result. Therefore, the detection and removal of empty orders in the disclosed embodiment can improve the accuracy of the scheduling result.

[0104] In a possible implementation, the method further includes: sorting the first orders executed in parallel in the scheduling result according to the order of receiving the orders, and updating the scheduling result to obtain an updated scheduling result.

[0105] In the disclosed embodiment, if the scheduling result contains at least two first orders with the same execution order (same start time and same end time) (i.e., first orders executed in parallel), the first orders can be reordered according to the order of receiving the orders to update the scheduling result. In this way, when personnel and equipment cannot support the parallel execution of multiple orders throughout the process, the updated scheduling result can not only resolve production conflicts but also meet the user's requirements for order time limits, thereby improving the consistency of the scheduling result with the actual production situation and order situation.

[0106] Figure 2 A flowchart of another packaging printing order scheduling method provided in an embodiment of the present disclosure.

[0107] In the disclosed embodiment, empty orders in multiple initial orders can be screened out to obtain multiple orders to be scheduled. Then, the execution time and lag time of each task in the multiple orders are determined, that is, the time constraint of each task is determined. And the resource consumption of each task is determined.

[0108] According to the default schedule, the resource consumption of these multiple orders can be traversed, the resource consumption of each order can be accumulated and the resource inventory balance can be determined. In the case of insufficient resource inventory balance, the user can decide whether to replenish the inventory. If the inventory is not replenished (that is, the user has selected the skip order operation at least once during the traversal process), the skipped order will be deleted from the default schedule to obtain the first schedule. If the user has selected to replenish the inventory every time during the traversal process, the default schedule will be used as the first schedule.

[0109] Furthermore, training samples can be prepared and used to train the initial model to obtain an order scheduling model. The method for preparing the training samples and the method for training the initial model have been introduced above and will not be repeated here. The time constraints, resource consumption, and the first schedule of each task of the multiple orders to be scheduled are input into the order scheduling model to obtain a scheduling result.

[0110] If there are first orders with the same execution order in the scheduling results, the order of the default scheduling can be referred to, and the first orders with the same order can be sorted according to the order of the default scheduling to obtain an updated scheduling result.

[0111] Figure 3 This is a schematic diagram of the structure of the packaging printing order scheduling system provided by the embodiment of the present disclosure. The system 20 includes:

[0112] A time constraint and resource consumption determination unit 21, used to obtain the time constraint of each task in a plurality of orders to be scheduled and the resource consumption of each task;

[0113] A first schedule determining unit 22, configured to adjust a default schedule based on resource inventory and resource consumption to obtain a first schedule;

[0114] The scheduling result determination unit 23 is used to determine a target scheduling algorithm based on the resource consumption, the first schedule and the time constraint using the order scheduling model, and to sort the multiple orders based on the target scheduling algorithm to obtain a scheduling result.

[0115] In a possible implementation, the time constraint and resource consumption determination unit 21 is further configured to:

[0116] Obtaining the personnel time, material time, and equipment time occupied by each of the tasks;

[0117] For a single task, determine the maximum value of material duration and equipment duration;

[0118] The sum of the maximum value and the personnel time of the single task is taken as the execution time of the single task to obtain the execution time of each task;

[0119] The time difference between the start time of each task and the start time of the previous task is obtained, and the time difference is used as the lag time of each task.

[0120] In a possible implementation, the first schedule determining unit 22 is further configured to:

[0121] According to the default schedule, traverse the resource consumption of the multiple orders, accumulate the resource consumption of each order and determine the resource inventory balance;

[0122] In the case where the resource inventory balance is insufficient, a new resource inventory balance is obtained in response to the received inventory replenishment amount, and the traversal is continued until the resource consumption of all orders is traversed;

[0123] In the case where the resource inventory surplus is insufficient, in response to the received skip order operation, the current order is screened out, and the traversal is continued until the resource consumption of all orders is traversed;

[0124] In the traversal, when only the inventory replenishment quantity is received, the default schedule is used as the first schedule;

[0125] During the traversal, when at least one skip order operation is received, the filtered orders are removed from the default schedule to obtain the first schedule.

[0126] In a possible implementation, the training process of the order scheduling model includes:

[0127] Acquire an initial model, wherein a plurality of scheduling algorithms are preset in the initial model;

[0128] Using the initial model, determining a preferred algorithm based on a single training sample;

[0129] Using an optimal algorithm to schedule the orders in the single training sample to obtain a first scheduling result;

[0130] Determining a similarity between the first scheduling result and a true value of the single training sample;

[0131] Determining a reward value according to the similarity;

[0132] According to the reward value, the probability values ​​corresponding to the multiple scheduling algorithms are adjusted until the similarity is not less than the similarity threshold or the multiple scheduling algorithms are all determined to be the preferred algorithm at least once, and the initial model is used to process other training samples; until the model convergence condition is met, the order scheduling model is obtained.

[0133] In a possible implementation, the training process of the order scheduling model further includes:

[0134] Determine the sample resource consumption of each sample task based on the sample personnel time, sample material time, and sample equipment time occupied by each sample task of the plurality of sample orders, as well as the personnel consumption per unit time, material consumption per unit time, and equipment consumption per unit time of each sample task;

[0135] Determining the sample execution time of each of the sample tasks based on each sample personnel time, sample material time, and sample equipment time;

[0136] The sample time difference between the start time of the sample task and the start time of the previous sample task is used as the sample lag time of the sample task, and the sample execution time and the sample lag time are used as the sample time constraint of the sample task;

[0137] The order of accepting orders for the sample tasks is used as a sample schedule, and the sample resource consumption, sample time constraints, and sample schedules corresponding to the multiple sample orders are used as a single training sample.

[0138] In a possible implementation, the system 20 further includes:

[0139] An initial order reading unit, used for reading a plurality of initial orders;

[0140] The to-be-scheduled order determination unit is used to detect the initial order and remove empty orders to obtain the multiple orders to be scheduled.

[0141] In a possible implementation, the system 20 further includes:

[0142] The scheduling result updating unit is used to sort the first orders executed in parallel in the scheduling result according to the order of receiving the orders, and update the scheduling result to obtain the updated scheduling result.

[0143] In some embodiments, the functions or modules included in the system provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0144] The embodiment of the present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and the computer program instructions implement the above method when executed by a processor. The computer-readable storage medium can be a volatile or non-volatile computer-readable storage medium.

[0145] An embodiment of the present disclosure further proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0146] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0147] Figure 4 The electronic device 1900 is provided as a server or a terminal device. Figure 4 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0148] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2003. TM , Mac OS X TM , Unix TM ,Linux TM , FreeBSD TM or similar.

[0149] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to perform the above method.

[0150] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0151] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0152] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0153] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0154] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0155] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0156] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0157] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0158] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A packaging printing order scheduling method, characterized in that: include: Obtaining the time constraints of each task in a plurality of orders to be scheduled and the resource consumption of each task; Based on the resource inventory and the resource consumption, adjusting the default schedule to obtain a first schedule; Based on the resource consumption, the first schedule and the time constraint, an order scheduling model is used to determine a target scheduling algorithm, and based on the target scheduling algorithm, the multiple orders are sorted to obtain a scheduling result.

2. The method according to claim 1, characterized in that The step of obtaining the time constraints of each task in the plurality of orders to be scheduled includes: Obtaining the personnel time, material time, and equipment time occupied by each of the tasks; For a single task, determine the maximum value of material duration and equipment duration; The sum of the maximum value and the personnel time of the single task is taken as the execution time of the single task to obtain the execution time of each task; The time difference between the start time of each task and the start time of the previous task is obtained, and the time difference is used as the lag time of each task.

3. The method according to claim 1, characterized in that: The adjusting the default schedule based on the resource inventory and the resource consumption to obtain the first schedule includes: According to the default schedule, traverse the resource consumption of the multiple orders, accumulate the resource consumption of each order and determine the resource inventory balance; In the case where the resource inventory balance is insufficient, a new resource inventory balance is obtained in response to the received inventory replenishment amount, and the traversal is continued until the resource consumption of all orders is traversed; In the case where the resource inventory surplus is insufficient, in response to the received skip order operation, the current order is screened out, and the traversal is continued until the resource consumption of all orders is traversed; In the traversal, when only the inventory replenishment quantity is received, the default schedule is used as the first schedule; During the traversal, when at least one skip order operation is received, the filtered orders are removed from the default schedule to obtain the first schedule.

4. The method according to claim 1, characterized in that The training process of the order scheduling model includes: Acquire an initial model, wherein a plurality of scheduling algorithms are preset in the initial model; Using the initial model, determining a preferred algorithm based on a single training sample; Using an optimal algorithm to schedule the orders in the single training sample to obtain a first scheduling result; Determining a similarity between the first scheduling result and a true value of the single training sample; Determining a reward value according to the similarity; According to the reward value, the probability values ​​corresponding to the multiple scheduling algorithms are adjusted until the similarity is not less than the similarity threshold or the multiple scheduling algorithms are all determined to be the preferred algorithm at least once, and the initial model is used to process other training samples; until the model convergence condition is met, the order scheduling model is obtained.

5. The method according to claim 4, characterized in that Also includes: Determine the sample resource consumption of each sample task based on the sample personnel time, sample material time, and sample equipment time occupied by each sample task of the plurality of sample orders, as well as the personnel consumption per unit time, material consumption per unit time, and equipment consumption per unit time of each sample task; Determining the sample execution time of each of the sample tasks based on each sample personnel time, sample material time, and sample equipment time; The sample time difference between the start time of the sample task and the start time of the previous sample task is used as the sample lag time of the sample task, and the sample execution time and the sample lag time are used as the sample time constraint of the sample task; The order of accepting orders for the sample tasks is used as a sample schedule, and the sample resource consumption, sample time constraints, and sample schedules corresponding to the multiple sample orders are used as a single training sample.

6. The method according to claim 1, characterized in that The method further comprises: Read multiple initial orders; The initial order is detected, and empty orders are removed to obtain the multiple orders to be scheduled.

7. The method according to claim 1, characterized in that The method further comprises: The first order executed in parallel in the scheduling result is sorted according to the order of receiving the order, and the scheduling result is updated to obtain an updated scheduling result.

8. A packaging printing order scheduling system, characterized in that: include: A time constraint and resource consumption determination unit, used to obtain the time constraint of each task in a plurality of orders to be scheduled and the resource consumption of each task; A first schedule determining unit, configured to adjust a default schedule based on resource inventory and resource consumption to obtain a first schedule; The scheduling result determination unit is used to determine a target scheduling algorithm based on the resource consumption, the first schedule and the time constraint using an order scheduling model, and to sort the multiple orders based on the target scheduling algorithm to obtain a scheduling result.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the method described in any one of claims 1 to 7 when executing the instructions stored in the memory.

10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images