Multi-Product Distributed Integrated Scheduling Method for Asymmetric Three-Workshops

Abstract

The present invention proposes a multi-product distributed integrated scheduling method for an asymmetric three-workshop system. The present invention takes "products" as the optimization object, designs an urgency value strategy and a multi-level feedback scheduling strategy. On the basis of following the layer priority principle, it fully considers the conflict problems of simultaneous processing of multiple products and the characteristics of urgent processes, such as their positions in complex product structures and processing times, to determine the scheduling order of processes; taking "workshops" as the optimization object, it designs a workshop assignment process strategy to reduce the number of migrations between different workshops, further shortening the time consumption of the entire processing process and improving the overall utilization rate of equipment. It takes into account both the characteristics of the "slender" tree structure with more levels and vertical extension of nodes, and the characteristics of the "flat" tree structure with fewer levels and more horizontal branches.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer integrated manufacturing, and particularly relates to a multi-product distributed integrated scheduling method for an asymmetric three-workshop Background Art

[0002] Traditional workshop scheduling mainly solves the pure processing problem where there is no sequential constraint relationship between workpieces, mainly including flow-shop scheduling problems and job-shop scheduling problems, etc. Many experts and scholars have carried out in-depth research on this. For example, for the large-scale multiple flexible job-shop scheduling problem with sequence-dependent setup times, a hybrid fluid master-apprentice evolutionary method is proposed to minimize the makespan; for the multi-objective flow-shop problem, an improved multi-objective grey wolf optimization method is proposed to improve the global search ability and local search ability of the algorithm; for the multi-objective zero-wait flow-shop scheduling problem, a mixed integer linear programming model is constructed, and a tensor-based evolutionary method accelerated by GPU is proposed to effectively solve this problem. Due to the continuous development of product demands towards personalization and diversification, there are more and more orders for customized and small-batch products. Different from traditional flow scheduling and workshop scheduling, a new scheduling mode is proposed, that is, an integrated scheduling that synchronously processes product processing and assembly.

[0003] Currently, domestic and foreign researchers' research on integrated scheduling problems mainly includes general integrated scheduling problems, flexible integrated scheduling problems, distributed integrated scheduling problems, multi-objective scheduling research, and multi-workshop integrated scheduling problems, etc. However, there are few research results on multi-workshop integrated scheduling problems. For example, for the integrated scheduling problem where the operation transfer time affects the product processing time, a two-workshop equipment-driven integrated scheduling method using the operation transfer time is proposed; for the multi-workshop collaborative integrated scheduling problem dealing with similar processing attributes, a hybrid teaching optimization method is proposed. However, these methods all consider how to allocate operations for processing when the equipment resources in the workshop are symmetric for a single product, ignoring the special cases of asymmetric equipment resources and simultaneous processing of multiple products. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems in the prior art, and a multi-product distributed integrated scheduling method for an asymmetric three-workshop is proposed. Specifically, in the distributed scheduling process of multiple complex products with a tree structure where "processing and assembly are carried out synchronously", taking the complex multi-product structure attributes and the asymmetric processing capabilities of multiple workshops as the dual optimization objects, an optimization effect of shorter total processing time for multiple products and higher overall utilization rate of equipment is achieved.

[0005] The present invention is realized through the following technical solutions. The present invention proposes a multi-product distributed integrated scheduling method for an asymmetric three-workshop, and the method includes the following steps:

[0006] Step 1: Form a virtual processing process tree from the processing process trees of multiple products;

[0007] Step 2: Calculate the urgency value P of each product according to the urgency value strategy i ;

[0008] Step 3: Adopt a multi-level feedback scheduling strategy to form a pre-scheduled operation set;

[0009] Step 4: Adopt a workshop assignment operation strategy to select the processing workshop of the operations in the pre-scheduled operation set;

[0010] Step 5: Store the operations with the determined processing workshops into the schedulable operation set, and delete the operations from the pre-scheduled operation set;

[0011] Step 6: Judge whether the pre-scheduled operation set is empty. If it is, go to step 7; if not, go to step 4;

[0012] Step 7: The scheduling ends and exits.

[0013] Furthermore, assume that the number of products is Z, the j-th operation of product i is U ij (1 ≤ i ≤ n, 1 ≤ j ≤ n), the processing workshop x of the product is W x (1 ≤ x ≤ n), the equipment resource sets of each workshop are M = {M x , 1 ≤ x ≤ n}, where the equipment resources of each workshop are asymmetric; let the starting time of processing product i be S i , TB ij , TE ij respectively represent the starting processing time and the completed processing time of the j-th operation of product i, and V represents the number of migrations of the processed product; under the condition of meeting the constraints, select the processable equipment of the appropriate processing workshop for each operation of all products to ensure that the number of migrations is as small as possible and the maximum completion time of all products is the smallest; thus, we can obtain:

[0014]

[0015] s.t. min TB ij (2)

[0016] TB ij ≥ S i (3)

[0017] TB (i+1)(j+1) ≥ max(TE ij ) (4)

[0018] min(V) (5)

[0019] In the formula: (1) represents minimizing the maximum completion time of the product, which is the optimization objective; (2) represents starting the processing of the operation as early as possible; (3) represents that the starting processing time of each operation shall not be earlier than the starting processing time of the product; (4) represents that the subsequent operation can only be processed after the completion of its preceding operation; (5) represents minimizing the number of migrations of operations during the scheduling process.

[0020] Furthermore, the urgency value strategy is specifically as follows:

[0021] The urgency value P of product i i is:

[0022]

[0023] In the formula: t ri represents the total processing duration of the urgent operations of product i; represents the sum of the processing times of all operations of product i; R i represents the total number of operations of product i; r i represents the number of urgent operations of product i.

[0024] Furthermore, the multi-level feedback scheduling strategy is specifically as follows:

[0025] Based on the number G ij of workshops where the operation can be processed, the operations at the same level in the virtual processing process tree are scheduled hierarchically. The operations with more workshops where they can be processed are assigned a higher priority level. The higher the level R, the earlier the processing; among operations at the same level, they are scheduled from the largest to the smallest according to the urgency value of the product. The larger the urgency value of the product, the earlier the processing; for operations with the same urgency value, they are scheduled from the smallest to the largest according to the short processing time strategy. The smaller the processing time of the product, the earlier the processing.

[0026] Furthermore, the workshop assignment operation strategy is specifically as follows: The operation needs to select a suitable processing device in the workshop for processing;

[0027] Case 1: When there is only one idle processing device for the operation in a workshop, two situations are considered:

[0028] If the processing device for the operation is in the same workshop as the processing device for its preceding operation, the operation does not need to be migrated and waits for the device in this workshop to be idle before processing, that is, in A i M x1k1 and A (i-1) M x2k2 among, then α = 0;

[0029] If the processing device for the operation is not in the same workshop as the processing device for its preceding operation with the latest processing end time, the operation needs to be migrated to the processing device in another workshop for processing, that is, Ai M x1k1 With A (i-1) M x2k2 Among them, then α = 1;

[0030] Case 2: When the equipment that can process the process is idle in multiple workshops, two cases are considered:

[0031] If the equipment that can process the process and the equipment that can process its immediate predecessor process are in the same workshop, the process is assigned to be processed in this workshop, that is, A i M x1k1 With A (i+1) M x2k2 Among them, then α = 0;

[0032] If the equipment that can process the process and the equipment that can process its immediate predecessor process are not in the same workshop, it is assigned to the equipment corresponding to the workshop with a smaller migration time, that is, A i M x1k1 With A (i+1) M x2k2 Among them, then α = 1, If the migration times are the same, it is assigned to the workshop where the process starts to be processed earliest.

[0033] Furthermore, the migration time of the process is equal to the ratio of the migration capabilities of the starting workshop and the migrated workshop multiplied by the processing time of the process, that is The migration capability refers to the speed at which the process in the processing workshop migrates to other processing workshops, denoted by λ x Indicates.

[0034] Furthermore, use α to represent the migration coefficient of the process. If it is 1, it means migration occurs, and if it is 0, it means no migration occurs.

[0035] Furthermore, what is used to measure the urgency of multiple products in the time dimension is the urgency value. The greater the urgency value of the product, the higher the processing urgency of the product.

[0036] The present invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the multi-product distributed integrated scheduling method for the asymmetric three workshops are implemented.

[0037] The present invention also provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the multi-product distributed integrated scheduling method for the asymmetric three workshops are implemented.

[0038] The beneficial effects of the present invention are:

[0039] (1) In the multi - product integrated scheduling, a multi - level feedback scheduling strategy is proposed. All processes are hierarchically scheduled, effectively solving the conflict problems generated during the co - processing of multi - products.

[0040] (2) In the multi - product integrated scheduling, the concept of product urgency value is introduced to determine the priority order of each process at the same level.

[0041] (3) For the integrated scheduling of the asymmetric three - workshop, a workshop - assigned process strategy is proposed, always maintaining the assignment principle of processing in the same workshop as the immediate - preceding process, reducing the number of migrations. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0043] Figure 1 It is the flowchart of the multi - product distributed integrated scheduling method for the asymmetric three - workshop described in the present invention;

[0044] Figure 2 It is the multi - level feedback scheduling strategy diagram;

[0045] Figure 3 It is the flowchart of the workshop - assigned process strategy;

[0046] Figure 4 It is the multi - workshop processing network diagram of Example 1;

[0047] Figure 5 It is the schematic diagram of the processing technology tree of Product A;

[0048] Figure 6 It is the schematic diagram of the processing technology tree of Product B;

[0049] Figure 7 It is the schematic diagram of the processing technology tree of Product C;

[0050] Figure 8 It is the schematic diagram of the virtual processing technology tree composed of Products A, B, and C;

[0051] Figure 9 It is the scheduling Gantt chart of Products A, B, and C;

[0052] Figure 10 It is the multi - workshop processing network diagram of Example 2;

[0053] Figure 11 It is the schematic diagram of the processing technology tree of Product P;

[0054] Figure 12 It is a schematic diagram of the processing process tree of product Q;

[0055] Figure 13 It is a schematic diagram of the processing process tree of product R;

[0056] Figure 14 It is a schematic diagram of the virtual processing process tree composed of products P, Q, and R;

[0057] Figure 15 It is a scheduling Gantt chart of products P, Q, and R. Specific implementation manners

[0058] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0059] Combined with Figures 1 - 15 , the present invention proposes a multi-product distributed integrated scheduling method for an asymmetric three-workshop, and the method includes the following steps:

[0060] step1: Form a virtual processing process tree from the processing process trees of multiple products;

[0061] step2: Calculate the urgency value P of each product according to the urgency value strategy i ;

[0062] Step3: Adopt a multi-level feedback scheduling strategy to form a pre-scheduled operation set;

[0063] Step4: Adopt a workshop assignment operation strategy to select the processing workshop of the operations in the pre-scheduled operation set;

[0064] Step5: Store the operations with the determined processing workshop into the schedulable operation set, and delete the operation from the pre-scheduled operation set;

[0065] Step6: Determine whether the pre-scheduled operation set is empty. If so, go to step7; otherwise, go to step4;

[0066] Step7: The scheduling ends and exits.

[0067] During the multi-product distributed integrated scheduling process of the asymmetric three-workshop, there are also some special constraint relationships between operations, and the specific requirements are as follows:

[0068] (1) If each process and its immediate predecessor process are not processed in the same workshop, then the process undergoes one migration. The more migrations a process has, the greater the migration time, and the longer the overall scheduling process will be.

[0069] (2) The equipment resources owned by asymmetric workshops are different, resulting in the fact that the processes on certain equipment can only be processed in fixed workshops.

[0070] For ease of description, the following related concepts are now defined:

[0071] Definition 1 Migration ability: The speed at which a process in a processing workshop migrates to another processing workshop, denoted by λ x The more equipment there is in a processing workshop, the greater the migration ability.

[0072] Definition 2 Migration time: The migration time of the process is equal to the ratio of the migration ability between the starting workshop and the workshop after migration multiplied by the processing time of the process, i.e.,

[0073] Definition 3 Migration coefficient: Denote the migration coefficient of the process by α. If it is 1, it means migration occurs; if it is 0, it means no migration occurs.

[0074] Definition 4 Urgency value: Used to measure the urgency of multiple products in the time dimension. The greater the urgency value of a product, the higher the processing urgency of the product.

[0075] Assume the number of products is Z, the j-th process of product i is U ij (1 ≤ i ≤ n, 1 ≤ j ≤ n), the processing workshop x of the product is W x (1 ≤ x ≤ n), the set of equipment resources of each workshop is M = {M x , 1 ≤ x ≤ n}, where the equipment resources of each workshop are asymmetric; let the starting processing time of product i be S i , TB ij , TE ij respectively represent the starting processing time and the completed processing time of the j-th process of product i, and V represents the number of migrations of the processed product; under the condition of meeting the constraints, select the processable equipment of a suitable processing workshop for each process of all products to ensure that the number of migrations is as small as possible and the maximum completion time of all products is minimized; thus, we can obtain:

[0076]

[0077] s.t.minTB ij (2)

[0078] TB ij ≥ S i (3)

[0079] TB(i+1)(j+1) ≥ max(TE ij ) (4)

[0080] min(V) (5)

[0081] In the formula: (1) represents minimizing the maximum completion time of the product, which is the optimization goal; (2) represents starting the processing of the process as early as possible; (3) represents that the starting processing time of each process shall not be earlier than the starting processing time of the product; (4) represents that the subsequent process needs to be processed after its preceding process is completed; (5) represents minimizing the number of migrations of the process during the scheduling process.

[0082] The present invention adopts a variety of strategies to optimize the scheduling process. First, a multi-level feedback scheduling strategy is adopted. By scheduling the processes in levels, the parallel processing ability of the equipment is significantly improved, and the optimization goal centered on the process is achieved; second, a product contribution value strategy is adopted, effectively reducing the processing time of parallel processes; finally, a workshop assignment process strategy is adopted, which is a strategy from the perspective of reducing the number of migrations and starting the processing of processes as early as possible, thereby increasing the intensity of compact processing of processes and the overall utilization rate of equipment.

[0083] Among them, the urgency value strategy is specifically as follows:

[0084] The urgency value P of product i i is:

[0085]

[0086] In the formula: t ri represents the total processing duration of the urgent processes of product i; represents the sum of the processing times of all processes of product i; R i represents the total number of processes of product i; r i represents the number of urgent processes of product i.

[0087] The multi-level feedback scheduling strategy is specifically as follows:

[0088] As Figure 2 shown, based on the number G ij of workshops where the process can be processed, the processes at the same level in the virtual processing process tree are scheduled in levels. The processes with more workshops where they can be processed are given a higher priority level. The higher the level R, the earlier the processing; the processes at the same level are scheduled according to the product urgency value from large to small. The larger the product urgency value, the earlier the processing; for processes with the same urgency value, they are scheduled from small to large according to the short processing time strategy. The smaller the product processing time, the earlier the processing.

[0089] As Figure 3 shown, the workshop assignment process strategy is specifically as follows: The process needs to select the processable equipment in a suitable workshop for processing;

[0090] Case 1: When the equipment available for processing a process is idle in only one workshop, two cases are considered:

[0091] If the equipment available for processing a process and the processing equipment of its immediate predecessor process are in the same workshop, the process does not need to be migrated and waits for the equipment in this workshop to be idle for processing, that is, A i M x1k1 and A (i-1) M x2k2 in, then α = 0;

[0092] If the equipment available for processing a process and the processing equipment of its immediate predecessor process with the latest processing end time are not in the same workshop, the process needs to be migrated to the equipment in other workshops for processing, that is, A i M x1k1 and A (i-1) M x2k2 in, then α = 1;

[0093] Case 2: When the equipment available for processing a process is idle in multiple workshops, two cases are considered:

[0094] If the equipment available for processing a process and the available processing equipment of its immediate predecessor process are in the same workshop, the process is assigned to this workshop for processing, that is, A i M x1k1 and A (i+1) M x2k2 in, then α = 0;

[0095] If the equipment available for processing a process and the available processing equipment of its immediate predecessor process are not in the same workshop, it is assigned to the equipment corresponding to the workshop with a smaller migration time for processing, that is, A i M x1k1 and A (i+1) M x2k2 in, then α = 1, If the migration times are the same, it is assigned to the workshop where the process starts processing earliest.

[0096] Method complexity analysis: The time complexity of calculating the number of urgent processes and the total duration of a product is O(n); the time complexity of calculating the urgency value of a product is O(n 2 ); the time complexity of forming a pre-scheduled process set through a multi-level feedback scheduling strategy is O(n); the time complexity of calling the workshop assignment process strategy is O(n 2 ); In summary, the time complexity of the method described in the present invention is max(O(n), O(n 2 )) = O(n 2 ).

[0097] For the multi - product distributed integrated scheduling method of the asymmetric three - workshop studied in the present invention, two different types of examples are selected for scheduling demonstration to comprehensively evaluate the performance and advantages of the method. Example 1 selects a complex product with an "elongated" tree structure, which has a relatively deep hierarchy. The nodes extend sequentially along the longitudinal depth direction, and the relationship between each level is close and distinct; Example 2 selects a complex product with a "flat" tree structure, which has a shallower hierarchy, has more branches in the horizontal dimension, and the nodes are widely distributed at the same level.

[0098] Each node in the figure includes three elements: the processing operation, the corresponding processing equipment, and the processing time required. The three elements are separated by " / ". The arrows in the figure represent the constraint relationships between adjacent operations during tight processing. The starting point of the arrow represents the preceding operation, and the direction of the arrow points to the succeeding operation.

[0099] Scheduling Analysis and Results of Example 1

[0100] Suppose there are multi - variety and small - batch products A, B, and C, and their processing process trees are as Figures 5 - 7 shown. And there are three asymmetric workshops, namely workshop a, workshop b, and workshop c. Among them, workshop a has equipment M2, M4, and M5, workshop b has equipment M1, M4, and M5, and workshop c has equipment M3 and M5, as Figure 4 shown.

[0101] Step1: Align products A, B, and C with the root nodes to form a virtual processing process tree, as Figure 8 shown.

[0102] Step2: Calculate the number of urgent operations, the total processing time, the total number of operations, and the total processing time of each product. According to the urgency value strategy, the urgency value of product A is 2.857, the urgency value of product B is 2.571, and the urgency value of product C is 1.5, as shown in Table 1. Therefore, the product scheduling order is A, B, C.

[0103] Table 1 Product Urgency Value Information Table

[0104]

[0105] Step3: For the Figure 8 virtual processing process tree, the levels of operations A8, A9, A10, and B11 are all 5. The multi - level feedback scheduling strategy is used for grading: among them, A8 and B11 are processed on equipment M5, A9 is processed on equipment M1, and A10 is processed on equipment M4. So G A8 = 3, G B11 = 3 is the first level, G A10 = 2 is the second level, G A9= 1 is the third level, and the urgency value of process A8 at the same level is greater than that of B11. Therefore, the scheduling order of processes at level 5 is {A8, B11, A10, A9}.

[0106] Step4: Add processes A8, B11, A10, and A9 to the pre-scheduling set. There are 7 processes at level 4, namely A6, A7, B8, B9, B10, C5, and C6. The multi-level feedback scheduling strategy is used for grading: G C5 = 3 is the first level, G A6 = 2, G B9 = 2, G C6 = 2 is the second level, G A7 = 1, G B8 = 1, G B10 = 1 is the third level, where the urgency value A > B > C. Therefore, the scheduling order of processes at level 4 is {C5, A6, B9, C6, A7, B8, B10}.

[0107] Step5: Similarly, all processes at levels 3, 2, and 1 are scheduled using the multi-level feedback scheduling strategy. Therefore, the scheduling order of processes at level 3 is {B7, B5, C4, A5, A3, A4, B4, B6, C3}; the scheduling order of processes at level 2 is {A2, C2, B2, B3}; the scheduling order of processes at level 1 is {C1, A1, B1}. Therefore, the pre-scheduling process set is {A8, B11, A10, A9, C5, A6, B9, C6, A7, B8, B10, B7, B5, C4, A5, A3, A4, B4, B6, C3, A2, C2, B2, B3, C1, A1, B1}.

[0108] Step6: According to the workshop allocation strategy, allocate workshops for the processes in the pre-scheduling process set: If process A8 and its immediate successor A6 and the immediate successor of the immediate successor A3 appear in the same workshop a for processing, then allocate it to workshop a for processing; if process B11 and its immediate successor B8 appear in the same workshop b for processing, then allocate it to workshop b; if process A10 and its immediate successor A7 do not appear in the same workshop for processing, and the transfer time It is then assigned to Workshop b, which is the earliest to start processing; Process A9 is processed on Equipment M1 and can only be assigned to Workshop a; if Process C5 and its immediate successor Process C3 are processed in the same workshop, it is assigned to Workshop c; if Process A6 and its immediate predecessor Process A8 are processed in the same Workshop a, it is assigned to Workshop a; if Process B9 and its immediate successor Process B5 are processed in the same Workshop a, it is assigned to Workshop a; if Process C6 and its immediate successor Process C4 are processed in multiple workshops, namely a and b, and the earliest processing time in Workshop b is earlier, it is assigned to Workshop b; A7 is processed on Equipment M3 and can only be assigned to Workshop c; if Process B8 and its immediate predecessor Process B11 are processed in the same Workshop b, it is assigned to Workshop b; Process B10 is processed on Equipment M2 and can only be assigned to Workshop a. Similarly, select the workshops where the remaining processes in the pre-scheduling can be processed, as shown in Table 2 specifically.

[0109] Table 2 Workshop Assignment Execution Table

[0110]

[0111]

[0112] To sum up, the total scheduling time for products A, B, and C is 120 man-hours, the number of migrations V is 6, and the scheduling Gantt chart is as Figure 9 shown.

[0113] Scheduling Analysis and Results of Example 2

[0114] Suppose there are multi-variety and small-batch products P, Q, and R, and their processing process trees are as Figures 11 - 13 shown. And there are three asymmetric workshops, namely Workshop e, Workshop f, and Workshop g. Among them, Workshop e has Equipment M1, Workshop f has Equipment M2, M3, and M4, and Workshop g has Equipment M2, M4, and M5, as Figure 10 shown.

[0115] Step1: Align products P, Q, and R with the root nodes to form a virtual processing process tree, as Figure 14 shown.

[0116] Step2: Calculate the number of urgent processes, total processing time, total number of processes, and total processing time of each product. According to the urgency value strategy, the urgency value of product A is 2.75, the urgency value of product B is 3.478, and the urgency value of product C is 1.455, as shown in Table 3. Therefore, the product scheduling order is Q, P, R.

[0117] Table 3 Product Urgency Value Information Table

[0118]

[0119] Step3: ForFigure 14 In the virtual machining process tree, the levels of processes P6, P7, P8, P9, Q10, Q11, Q12, and R6 are all 4, and a multi-level feedback scheduling strategy is used for grading: among them, G P6 = 2, G P9 = 2, G Q10 = 2, G Q11 = 2 is the first level, G P7 = 1, G P8 = 1, G Q12 = 1, G R6 = 1 is the second level. Among them, the urgency values Q > P > R, so the scheduling order of processes with level 4 is {Q11, Q10, P6, P9, Q12, P7, P8, R6}.

[0120] Step4: Add processes Q11, Q10, P6, P9, Q12, P7, P8, and R6 to the pre-scheduling set. There are 10 processes with level 3, namely P4, P5, Q5, Q6, Q7, Q8, Q9, R3, R4, and R5, and a multi-level feedback scheduling strategy is used for grading: G Q5 = 2, G Q8 = 2, G Q7 = 2, G P5 = 2, G P4 = 2, G R3 = 2 is the first level, G Q6 = 1, G Q9 = 1, G R4 = 1, G R5 = 1 is the second level. Among them, the urgency values Q > P > R, so the scheduling order of processes with level 3 is {Q5, Q8, Q7, P5, P4, R3, Q6, Q9, R5, R4}.

[0121] Step5: Similarly, all processes with levels 2 and 1 are scheduled using a multi-level feedback scheduling strategy: the scheduling order of processes with level 2 is {Q3, Q4, P3, R2, Q2, P2}; the scheduling order of processes with level 1 is {R1, Q1, P1}. Therefore, the pre-scheduling process set is {Q11, Q10, P6, P9, Q12, P7, P8, R6, Q5, Q8, Q7, P5, P4, R3, Q6, Q9, R5, R4, Q3, Q4, P3, R2, Q2, P2, R1, Q1, P1}.

[0122] Step 6: According to the workshop allocation strategy, allocate workshops for the processes in the pre-scheduled process set: For process Q11 and its succeeding process Q7, there is a situation where they can be processed in multiple workshops f and g, and another preceding process Q12 of process Q7 can only be processed in workshop g. To reduce the number of migrations, Q11 is allocated to workshop g for processing; for process Q10 and its succeeding process Q5, there is a situation where they can be processed in multiple workshops f and g. Since the earliest start processing time is in workshop f, Q10 is allocated to workshop f; for process P6 and its succeeding process P4, there is a situation where they can be processed in multiple workshops f and g, and another preceding process P7 of process P4 can only be processed in workshop f, then P6 is allocated to workshop f; for process P9 and its succeeding process P5, there is a situation where they can be processed in multiple workshops f and g. Since the earliest start processing time is in workshop g, Q10 is allocated to workshop g; process Q12 is processed on equipment M5, so it is only allocated to workshop g; process P7 is processed on equipment M3, so it is only allocated to workshop f; process P8 is processed on equipment M1, so it is only allocated to workshop e; process R6 is processed on equipment M3, so it is only allocated to workshop f; for process Q5 and its preceding process Q10, they are processed in the same workshop f, so it can be allocated to workshop f; for process Q8 and its succeeding process Q4, there is a situation where they can be processed in multiple workshops f and g, and the earliest start processing time is in workshop f, then Q8 is allocated to workshop f; for process Q7 and its preceding processes Q11, Q12, they are processed in the same workshop g, then Q7 is allocated to workshop g. Similarly, select the processable workshops for the remaining processes in the pre-scheduling, as shown in Table 4 specifically.

[0123] Table 4 Workshop Allocation Execution Table

[0124]

[0125]

[0126] To sum up, the total scheduling time of products P, Q, and R is 130 man-hours, the number of migrations V is 6, and the scheduling Gantt chart is as Figure 15 shown.

[0127] The main advantages of the multi-product distributed integrated scheduling method for an asymmetric three-workshop proposed by the present invention are as follows:

[0128] (1) From the perspective of optimizing the realization of compact process scheduling and improving the overall utilization rate of equipment, the urgency value strategy and the multi-level feedback scheduling strategy are adopted to determine the scheduling order of processes, effectively shortening the equipment idle period and the parallel processing time of products, and achieving the fast scheduling effect of close processing of processes on equipment. Taking Figure 9 as an example, equipment W b M1 is closely processed from t = 0 to t = 120, and its equipment utilization rate reaches 100%; taking Figure 15 as an example, equipment W eM1 is closely processed from t = 0 to t = 110, and its equipment utilization rate reaches 100%, as shown in Table 5 specifically.

[0129] Table 5 Scheduling Results and Equipment Utilization of Two Examples

[0130] Example Total processing time for multiple products Overall equipment utilization rate 1 120 87.1％ 2 130 80.6％

[0131] (2) From the perspective of optimizing by reducing the number of migrations and shortening the processing time, the present invention adopts a workshop allocation process strategy to reasonably allocate processing workshops for processes. During scheduling, the immediate predecessor process and the process are always processed in the same workshop, which greatly reduces the migration phenomenon and the number of migrations between different workshops, and thus realizes a more efficient resource allocation and significantly shortens the total processing time of products. Taking Figure 9 as an example, the total processing time of products A, B, and C is 120 man-hours, and the number of migrations V is 6; taking Figure 15 as an example, the total processing time of products P, Q, and R is 130 man-hours, and the number of migrations V is 6.

[0132] The present invention proposes a multi-product distributed integrated scheduling method for an asymmetric three-workshop: taking "products" as the optimization object, a urgency value strategy and a multi-level feedback scheduling strategy are designed. On the basis of following the layer priority principle, the conflict problem of simultaneous processing of multiple products and the characteristics such as the position and processing time of urgent processes in complex product structures are fully considered to determine the scheduling order of processes; taking "workshops" as the optimization object, a workshop allocation process strategy is designed to reduce the number of migrations between different workshops, further shortening the time consumption of the entire processing process and improving the overall utilization rate of equipment.

[0133] The present invention takes into account both the "slender" tree structure characteristics with more levels and vertical extension of nodes and the "flat" tree structure characteristics with fewer levels and more horizontal branches. The experimental results show that the present invention is applicable to different types of complex product structures, can not only ensure the minimization of the total processing time of multi-product processing during distributed scheduling, but also achieve the dual optimization effect of "products + workshops" to the greatest extent.

[0134] The present invention also proposes an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the multi-product distributed integrated scheduling method for the asymmetric three-workshop are realized.

[0135] The present invention also proposes a computer-readable storage medium for storing computer instructions, and when the computer instructions are executed by a processor, the steps of the multi-product distributed integrated scheduling method for the asymmetric three-workshop are realized.

[0136] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include but not limited to these and any other suitable types of memory.

[0137] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a high-density digital video disc (DVD)), or a semiconductor medium (such as a solid state disc (SSD)), etc.

[0138] In the implementation process, the steps of the above method can be completed by the integrated logic circuit of the hardware in the processor or the instructions in the form of software. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed by the hardware processor or executed by the combination of the hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0139] It should be noted that the processor in the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above method embodiments can be completed by the integrated logic circuit in the hardware of the processor or instructions in software form. The above-mentioned processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by a combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0140] The above has introduced in detail the multi-product distributed integrated scheduling method for the asymmetric three workshops proposed by the present invention. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. Asymmetric three-workshop multi-product distributed integrated scheduling method, characterized in that: The method comprises the following steps: Step 1: Form a virtual processing tree from the processing technology trees of multiple products; Step 2: Calculate the urgency value P of each product according to the urgency value strategy i ; Step 3: Adopt a multi-level feedback scheduling strategy to form a set of pre-scheduling processes; Step 4: Use the workshop allocation process strategy to select the processing workshop of the process in the pre-scheduled process set; Step 5: Store the process of the determined processing workshop into the set of schedulable processes, and delete the process from the set of pre-scheduled processes; Step 6: Determine whether the pre-scheduled process set is empty. If yes, go to step 7; if not, go to step 4. Step7: Scheduling is completed and exit; Assume that the number of products is Z and the jth process of product i is U ij (1≤i≤n, 1≤j≤n), the product's processing workshop x is W x (1≤x≤n), the equipment resource set of each workshop is M={M x , 1≤x≤n}, where the equipment resources of each workshop are asymmetric; let the processing start time of product i be S i , TB ij TE ij Respectively represent the starting processing time and the completion processing time of the jth process of product i, and V represents the number of migrations of the processed product; under the condition of satisfying the constraints, select the processable equipment of the appropriate processing workshop for each process of all products to ensure that the number of migrations is as small as possible and the maximum completion time of all products is minimized; thus, it can be obtained: s.t.minTB ij (2) TB ij ≥S i (3) TB (i+1)(j+1) ≥max(TE ij ) (4) min(V)(5) In the formula: (1) represents minimizing the maximum completion time of the product, which is the optimization goal; (2) represents starting the process as early as possible; (3) represents that the starting processing time of each process must not be earlier than the starting processing time of the product; (4) represents that the subsequent process can only be processed after the processing of the previous process is completed; (5) represents that the number of process migrations during the scheduling process is as small as possible; α represents the migration coefficient of the process, if it is 1, it means migration occurs, if it is 0, it means no migration occurs; λ x1 represents the migration capacity of the starting workshop, λ x2 represents the migration capacity of the workshop after migration, t ij Indicates the processing time of a certain process.

2. The method according to claim 1, characterized in that The urgency value strategy is specifically: The urgency value P of product i i for: Where: t ri represents the total processing time of the urgent process of product i; Represents the sum of the processing time of all processes of product i; R i represents the total number of processes for product i; r i Represents the number of urgent processes for product i.

3. The method according to claim 2, characterized in that The multi-level feedback scheduling strategy is specifically: The number of workshops that can process the process G ij As a benchmark, the processes at the same level in the virtual processing technology tree are hierarchically scheduled. The processes with more processing workshops are assigned higher priority levels. The higher the level R, the higher the priority for processing. The processes at the same level are scheduled from large to small according to the product urgency value. The larger the product urgency value, the higher the priority for processing. The processes with the same urgency value are scheduled from small to large according to the short-time strategy. The shorter the product processing time, the higher the priority for processing.

4. The method according to claim 3, characterized in that The workshop allocation process strategy is specifically as follows: the process needs to select the processable equipment in the appropriate workshop for processing; Case 1: When the equipment that can be processed in the process is idle in only one workshop, two cases are considered: If the process equipment is in the same workshop as the processing equipment of the previous process, the process does not need to be moved and can only be processed when the equipment in this workshop is idle. and middle, Then α=0; If the processing equipment of a process is not in the same workshop as the processing equipment of the process immediately before it with the latest processing completion time, the process needs to be moved to the equipment of another workshop for processing. and middle, Then α=1; Case 2: When the equipment that can be processed in the process is idle in multiple workshops, two cases are considered: If the processable equipment of a process and the processable equipment of its immediate preceding process are in the same workshop, the process will be assigned to this workshop for processing. and middle, Then α=0; If the process equipment is not in the same workshop as the process equipment immediately before it, it will be allocated to the equipment in the workshop with the shorter migration time. and middle, Then α=1, If the migration time is the same, it will be allocated to the workshop where the process starts earliest.

5. The method according to claim 4, characterized in that The migration time of a process is equal to the ratio of the migration capacity of the starting workshop to that of the migrated workshop multiplied by the processing time of the process, that is, Migration capability refers to the speed at which processes in a processing workshop can be migrated to other processing workshops, expressed as λ x express.

6. The method according to claim 5, characterized in that The urgency value is used to measure the urgency of multiple products in the time dimension. The larger the urgency value of a product, the more urgent the processing of the product.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images