Blockage flow shop scheduling method based on double-layer production-maintenance iterative optimization

Abstract

The invention relates to a blocking flow shop scheduling method based on double-layer production-maintenance iterative optimization. The method comprises the following steps: solving a multi-target blocking flow shop scheduling model through diverse search and enhanced search in sequence; the diversified search and the enhanced search are respectively a production-maintenance iterative optimization mode based on a processing sequence and a processing fragment granularity variable neighborhood search, and the production-maintenance iterative optimization mode comprises a production scheduling optimization-maintenance activity following module and a maintenance plan optimization-production activity following module. The method comprises the following steps: performing variable neighborhood search through a local search operator to obtain a processing sequence, and generating a corresponding maintenance sequence according to a machine minimum reliability principle; and the following steps are repeated for multiple times: adjusting the maintenance sequence according to the forward adjustment condition, and adjusting the processing sequence. Compared with the prior art, maintenance information can be effectively embedded into a scheduling decision, and a scientific integrated scheduling scheme is efficiently obtained.

Description

Technical Field

[0001] The invention belongs to the technical field of workshop production scheduling, and in particular relates to a blocking flow workshop scheduling method based on double-layer production-maintenance iterative optimization. Background Art

[0002] Assembly shop production is an efficient production method widely used in the manufacturing industry, which requires that all operations must pass through each machine in the same order. However, the particularity of industries such as steel, chemicals, and robotics often leads to a lack of buffers between adjacent machines. Space limitations or technical requirements mean that when downstream machines are busy, the completed operations of upstream machines cannot be directly transferred to the next step, resulting in the current process being blocked. The resulting blocked assembly shop scheduling problem is more complex than the traditional assembly line scheduling problem.

[0003] In this complex production environment, the segmented local optimization ideas often cannot meet the global needs, and the introduction of a comprehensive perspective has become a key means to solve complex production problems. Chinese invention patent CN113902174B discloses an improved sparrow search optimization method for the scheduling problem of a blocked flow shop. In the process of searching and updating the population position of the sparrow, an elite solution set and a multi-group dynamic learning strategy are introduced to improve the adequacy of the global search and balance the local mining and global exploration capabilities of the sparrow algorithm. However, in order to improve the overall competitiveness, enterprises not only pursue reducing production costs and improving efficiency, but also pay more attention to production safety and effective management of resources. In this context, the close connection and collaborative optimization between production scheduling and equipment maintenance activities have gradually become the research focus of the scheduling problem of blocked flow shops. By jointly optimizing production-maintenance, the resource utilization rate of equipment can be improved, unnecessary downtime can be reduced, and ultimately the operating cost can be reduced. However, the scheduling problem of a blocked flow shop considering maintenance activities is extremely complex in terms of constraints. Although some studies have begun to explore the combination of production and maintenance, the exploration of collaborative optimization models is still in its initial stage. The existing methods are insufficient in multi-objective optimization based on collaborative optimization scheduling, and it is difficult to meet the dual requirements of modern manufacturing industry for improving efficiency and optimizing management. Therefore, it is necessary to design a scheduling method for a blocked flow shop to effectively embed maintenance information into scheduling decisions, efficiently obtain scientific integrated scheduling solutions, and provide innovative solutions for enterprises to improve productivity and increase added value. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a blocked flow shop scheduling method based on two-layer production-maintenance iterative optimization, which effectively embeds maintenance information into scheduling decisions and efficiently obtains a scientific integrated scheduling solution.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] The present invention provides a blocking flow shop scheduling method based on double-layer production-maintenance iterative optimization, comprising the following steps:

[0007] Obtain basic equipment information and equipment status data, input the multi-objective blocked flow shop scheduling model, solve and obtain the optimal production-maintenance integrated scheduling solution, and schedule the blocked flow shop;

[0008] Among them, the production-maintenance integrated scheduling scheme includes a production sequence and a maintenance sequence, and the basic equipment information includes the time consumed by each machine to process different workpieces, as well as the processing threshold and maintenance time of each machine; the multi-objective blocking flow shop scheduling model aims to minimize the completion time and total cost, and is solved in turn by diverse search and enhanced search; the diverse search and the enhanced search are production-maintenance iterative optimization modes based on variable neighborhood search of processing sequence granularity and variable neighborhood search of processing fragment granularity, respectively, and the production-maintenance iterative optimization mode includes a production scheduling optimization-maintenance activity following module and a maintenance plan optimization-production activity following module. In the production scheduling optimization-maintenance activity following module, the following steps are repeated multiple times: a variable neighborhood search is performed through a local search operator to obtain a processing sequence, and a corresponding maintenance sequence is generated according to the principle of minimum reliability of the machine; in the maintenance plan optimization-production activity following module, the following steps are repeated multiple times: the maintenance sequence generated by the production scheduling optimization-maintenance activity following module is adjusted according to the forward adjustment condition, and the processing sequence generated by the production scheduling optimization-maintenance activity following module is adjusted.

[0009] Furthermore, the local search operator is any one of swap, insert and reverse.

[0010] Furthermore, the principle of minimum machine reliability is that before processing a workpiece, the cumulative processing time of the machine that processes the workpiece after completing the processing task of the workpiece is calculated. If it exceeds the processing threshold of the machine, the machine needs to be maintained before processing the workpiece; after the maintenance operation is completed, the machine continues to process.

[0011] Furthermore, the specific process of adjusting the maintenance sequence generated by the production scheduling optimization-maintenance activity following module according to the forward adjustment conditions is: if moving a maintenance operation forward does not increase the overall number of maintenance operations, and the previous idle time of the corresponding machine is sufficient to arrange the maintenance activity, then the maintenance operation and all production and maintenance operations after the maintenance operation will be moved forward as a whole.

[0012] Furthermore, the specific process of solving the multi-objective blocking flow shop scheduling model and obtaining the optimal production-maintenance integrated scheduling solution is as follows:

[0013] S1, randomly generate a processing sequence and set the number of iterations to 1;

[0014] S2, performing variable neighborhood search on the complete processing sequence through a local search operator to obtain a first processing sequence, and generating a corresponding first maintenance sequence according to the principle of minimum reliability of the machine;

[0015] S3, adjusting the first maintenance sequence according to the forward adjustment condition, and then adjusting the first processing sequence, and judging whether the first termination condition is met, if not, returning to step S2, if yes, proceeding to step S4;

[0016] S4. Obtain multiple processing segments according to the first maintenance sequence and the first processing sequence, perform variable neighborhood search on the processing segments through a local search operator to obtain a second processing sequence, and generate a corresponding second maintenance sequence according to a principle of minimum reliability of the machine;

[0017] S5, adjusting the second maintenance sequence according to the forward adjustment condition, and then adjusting the second processing sequence, and judging whether the second termination condition is met, if not, returning to step S4, if yes, proceeding to step S6;

[0018] S6. Generate a production-maintenance integrated scheduling plan based on the second processing sequence and the second maintenance sequence, and determine whether the iteration termination condition is met. If not, add 1 to the number of iterations and return to step S2; if so, output the current production-maintenance integrated scheduling plan, which is the optimal production-maintenance integrated scheduling plan.

[0019] Furthermore, the first termination condition and the second termination condition are both reaching a preset number of executions, and the iteration termination condition is reaching a preset number of iterations.

[0020] Furthermore, according to the first maintenance sequence and the first processing sequence, a specific process of obtaining a plurality of processing segments is as follows:

[0021] implement in, represents the integrated production-maintenance scheduling solution, ⊙ represents the first maintenance sequence Insert 0 at the corresponding position of the first processing sequence λ;

[0022] Taking 0 as the node, the production-maintenance integrated scheduling solution Divide into multiple processing segments.

[0023] Furthermore, the constraints of the multi-objective blocking flow shop scheduling model include:

[0024] The workpieces to be processed must be processed in the same order by all machines;

[0025] Workpieces that have completed the current operation must remain on the current machine until the next machine is free;

[0026] A machine can only process one workpiece at a time;

[0027] A workpiece can only be processed on one machine at a time, and no interruptions are allowed;

[0028] All machines are considered for maintenance to ensure a minimum reliability level, i.e. maintenance activities are performed on the corresponding machines before the accumulated processing time of each machine exceeds a threshold.

[0029] Furthermore, the specific expression of the constraint conditions of the multi-objective blocking flow shop scheduling model is as follows:

[0030]

[0031]

[0032] D k,1 ≥D (k-1),1 +p k,1 ,k=2,…,n

[0033]

[0034] Where n and m represent the number of workpieces and machines respectively, i and z represent the corresponding indexes respectively; k represents the position of the workpiece in the processing sequence; X i,k is a binary variable, when workpiece J i When it is at the kth position in the processing sequence, its value is 1, otherwise it is 0; k,z is a binary variable. When machine z is maintained after the kth workpiece is processed, its value is 1, otherwise it is 0. max Indicates completion time; S k,z With D k,z They represent the start and departure time of the kth workpiece on machine z; r z represents the time required for machine z to perform maintenance, p i,z represents the processing time of workpiece i on machine z; d z represents the processing threshold of machine z, G k,z represents the cumulative processing time of machine z to process the kth workpiece. When the machine is maintained, G k,z Reset; E k,z It represents the end time of the maintenance activity after processing the kth workpiece on machine z.

[0035] Furthermore, the total cost includes maintenance cost and machine operation cost.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention proposes a blocking flow shop scheduling method based on double-layer production-maintenance iterative optimization. With the goal of minimizing the completion time and total cost, a multi-objective blocking flow shop scheduling model is established to solve and obtain the optimal production-maintenance integrated scheduling solution; the model is solved in turn by diverse search and enhanced search, and the diverse search and enhanced search are production-maintenance iterative optimization modes based on variable neighborhood search of processing sequence granularity and variable neighborhood search of processing fragment granularity, respectively. Through variable neighborhood search of different granularities, the depth and breadth of the search can be balanced, the problem of local optimal solution can be overcome, and a solution closer to the global optimal solution can be found; the production-maintenance iterative optimization mode includes a production scheduling optimization-maintenance activity following module and a maintenance plan optimization-production activity following module. In the production scheduling optimization-maintenance activity following module, multiple Repeat the following steps many times: perform variable neighborhood search through local search operators to obtain processing sequences, and generate corresponding maintenance sequences according to the principle of minimum machine reliability. The above process can effectively insert maintenance activities while optimizing the production scheduling plan, thereby optimizing the production scheduling plan and following maintenance activities. In the maintenance plan optimization-production activity following module, repeat the following steps many times: adjust the maintenance sequence generated by the production scheduling optimization-maintenance activity following module according to the forward adjustment conditions, and adjust the processing sequence generated by the production scheduling optimization-maintenance activity following module. The above process can further adjust the maintenance activities, and then adjust the production activities as a whole, thereby updating the maintenance plan and following the production activities. The above method can obtain high-quality scheduling solutions in a relatively short time, and can provide decision makers with a variety of scheduling plans with different preferences, thereby enhancing the comprehensive competitiveness of the enterprise.

[0038] 2. At the beginning of solving the model, the present invention randomly generates a processing sequence, which helps to cover different areas of the solution space and maintain the diversity of the population. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the method of the present invention;

[0040] Figure 2 It is a schematic diagram of the structure of the production-maintenance iterative optimization model;

[0041] Figure 3 A schematic diagram for dividing the processing sequence segments;

[0042] Figure 4 is the Gantt chart corresponding to a set of integrated scheduling solutions in the example.

[0043] Among them, Processing time represents processing time, Blocking time represents blocking time, Maintenance time represents maintenance time, J and M represent workpiece and machine respectively;

[0044] Figure 5 It is a comparison chart of different indicators of the method of the present invention and the existing method,

[0045] Among them, (5a) corresponds to the convergence index, (5b) corresponds to the diversity index, and (5c) corresponds to the dominance index. DETAILED DESCRIPTION

[0046] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0047] Example:

[0048] This embodiment needs to solve the multi-objective blocked flow shop scheduling problem, with the goal of minimizing the completion time and total cost, and establish a multi-objective blocked flow shop scheduling model. The specific expressions of the model and its constraints are respectively as shown in formula (1) and formula (2)-formula (12):

[0049] {C max ,TC}(1)

[0050]

[0051] D k,1 ≥D (k-1),1 +p k,1 ,k=2,…,n (8)

[0052]

[0053] Among them, C max represents the completion time, TC represents the total cost (including maintenance cost and machine operation cost). n and m represent the number of workpieces and machines respectively, i and z represent the corresponding indexes respectively; k represents the position of the workpiece in the processing sequence; X i,k is a binary variable, when workpiece J i When it is at the kth position in the processing sequence, its value is 1, otherwise it is 0; k,z is a binary variable. When machine z is maintained after the kth workpiece is processed, its value is 1, otherwise it is 0. max Indicates completion time; S k,z With D k,z They represent the start and departure time of the kth workpiece on machine z; r z represents the time required for machine z to undergo maintenance, p i,z represents the processing time of workpiece i on machine z; d z represents the processing threshold of machine z, G k,zrepresents the cumulative processing time of machine z to process the kth workpiece. When the machine is maintained, G k,z Reset; E k,z It represents the end time of the maintenance activity after processing the kth workpiece on machine z.

[0054] Formulas (2) and (3) are used to ensure that each workpiece must be assigned to the processing sequence; Formula (4) shows that the completion time is greater than or equal to the time for each workpiece to complete processing; Formula (5) shows that the start time of each workpiece on the first machine must be greater than 0; Formulas (6) and (7) ensure that machine z cannot be started to process the kth workpiece before the (k-1)th maintenance activity on machine z is completed; Formulas (8) and (9) describe the time when the kth workpiece leaves the machine; Formula (10) shows the completion time of each job; Formula (11) ensures that the cumulative processing time of each machine does not exceed the corresponding threshold; Formula (12) stipulates that the maintenance activity needs to be performed after the current workpiece is processed.

[0055] This embodiment provides a blocking flow shop scheduling method based on two-layer production-maintenance iterative optimization. The overall idea is to optimize a specific business and follow up another business as a single business optimization approach, such as Figure 2 As shown in the figure, the optimization of production and maintenance is carried out in a rolling iteration, including:

[0056] 1) Production scheduling optimization - maintenance activity tracking: Access equipment status data and basic equipment information to obtain the latest maintenance time of the equipment. Convert the equipment maintenance requirements reflected in the basic equipment information into production scheduling constraints, and while optimizing the production scheduling plan, enable maintenance activities to be effectively inserted to obtain an integrated scheduling plan.

[0057] 2) Maintenance plan optimization - production activity following: Under the premise of keeping the number of maintenance activities unchanged, adjust the operation time of maintenance activities according to the machine downtime, and then adjust the production schedule to update the integrated scheduling plan.

[0058] At the same time, on the basis of the production-maintenance iterative optimization mode formed by the combination of 1) and 2), two variable neighborhood searches with different search granularities are introduced to construct diversified search and enhanced search, thereby forming a two-layer optimization method to effectively solve the problem to be solved by the present invention.

[0059] The method of this embodiment is as follows Figure 1 As shown, the specific steps include:

[0060] Step 1: Initialize the population. Randomly generate individuals of length n, which can be expressed as λ=(λ1,λ2,…,λ n ), where λ i(i=1,…,n) represents the i-th workpiece processed. The population capacity is PS. This random generation method helps to cover different areas of the solution space and maintain the diversity of the population.

[0061] Decode the workpieces in order and calculate the fitness value (i.e., the objective function value) of each individual. Calculate the cumulative processing time of each machine and obtain the corresponding maintenance node according to the processing threshold. Since the maintenance node of each machine may be different, the two-dimensional sequence of the maintenance plan will be obtained after decoding. in It means that machine z is maintained after processing the i-th workpiece, otherwise it is 0. At the same time, let the number of iterations gen = 1.

[0062] Step 2: Update all individuals through diversity search, according to Figure 2 The production-maintenance iterative optimization mode shown is performed until a preset number of executions is reached.

[0063] Step 2.1: Randomly select one of the three local search operators swap, insert, and reverse, perform local search on the complete processing sequence, and obtain the first processing sequence. Then, convert the equipment maintenance requirements into production scheduling constraints, and obtain the first maintenance sequence based on the principle of minimum machine reliability, so that maintenance activities can be effectively followed while optimizing the production scheduling plan. The principle of minimum machine reliability is that before processing a workpiece, the cumulative processing time of the machine that processes the workpiece after completing the processing task of the workpiece is calculated. If it exceeds the processing threshold of the machine, the machine must be maintained before processing the workpiece; after the maintenance operation is completed, the machine continues to process. The principle of minimum machine reliability can ensure that the cumulative processing time of the machine does not exceed the processing threshold and that the workpiece processing process will not be interrupted.

[0064] Step 2.2: Adjust the first maintenance sequence while keeping the number of maintenance activities unchanged and the start time of the current workpiece unchanged. According to the downtime of the machine, adjust the maintenance activities that meet the forward adjustment conditions (the forward movement of the activity does not increase the number of overall maintenance activities, and the idle time of the previous machine is sufficient to arrange the maintenance activity). Then, move all activities after the workpiece originally corresponding to the maintenance activity forward as a whole to adjust the first processing sequence.

[0065] By rolling out steps 2.1 and 2.2, the maintenance plan is updated and production activities follow.

[0066] Step 3: Update all individuals through enhanced search. Similarly, update according to Figure 2 The production-maintenance iterative optimization mode shown is performed until a preset number of executions is reached.

[0067] Step 3.1: First, divide the processing segments. Figure 3 As shown in Figure 1, the process segmentation is determined by inserting the maintenance activities of each machine in a production sequence. in, represents an integrated sequence of artifacts and maintenance activities, and ⊙ represents a maintenance schedule based on each device Insert 0 at the corresponding position of the processing sequence λ. For example, it contains four processing segments: [4,8,1,9], [10,5,6], [3] and [7,2]. Then repeat steps 3.1 and 3.2 multiple times.

[0068] Step 3.2: Randomly select one of the three local search operators, swap, insert, and reverse, and perform local search on the random processing fragment to obtain the second processing sequence. Then, convert the equipment maintenance requirements into production scheduling constraints, obtain the maintenance plan based on the principle of minimum machine reliability, and obtain the second maintenance sequence, so that the maintenance activities can be effectively followed while optimizing the production scheduling plan.

[0069] Step 3.3: Under the premise of keeping the number of maintenance activities unchanged and the start time of the current workpiece unchanged, adjust the maintenance activities that meet the conditions forward according to the machine's downtime, adjust the second maintenance sequence, and then adjust the production activities as a whole and adjust the second processing sequence.

[0070] By rolling out steps 3.1 and 3.2, the maintenance plan is updated and production activities follow.

[0071] Step 4: Get the current non-dominated solution set and update the global Pareto archive set. Determine whether the termination condition is met (for example, the preset number of iterations is met). If the termination condition is met, output the global Pareto archive set as the final result; otherwise, gen = gen + 1, and return to step 2.

[0072] To verify the effectiveness of the above method, this example uses 20 workpieces (J1, J2, ..., J 20 ), 5 machines (M1, M2, …, M5) are used for verification. The cumulative processing time threshold of each machine is 300, and the maintenance activity takes 18 minutes each time. The specific processing schedule of each workpiece is shown in Table 1:

[0073] Table 1 Basic information of equipment in the example of the embodiment

[0074]

[0075] Figure 4The Gantt chart corresponding to a set of integrated scheduling solutions for this instance is shown in Figure 2. The above method is compared with the classic and excellent multi-objective optimization algorithms NSGA2 and SPEA2, and evaluated from the three perspectives of convergence, diversity, and dominance. The smaller the convergence and diversity index values, the better, and the larger the dominance index value, the better.

[0076] For the above example, the three algorithms were run 20 times independently at 2.5 seconds, 5 seconds, and 10 seconds respectively. The interval diagrams of the results obtained by each algorithm are shown in Figure 2. Figure 5 As shown, (5a) corresponds to the convergence index, (5b) corresponds to the diversity index, and (5c) corresponds to the dominance index. Obviously, the method proposed in this embodiment can more effectively solve the blocking flow shop scheduling problem considering maintenance.

[0077] If the above method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0078] The above description of the embodiments is to facilitate the understanding and use of the invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. A blocking flow shop scheduling method based on two-layer production-maintenance iterative optimization, characterized in that: The following steps are involved: Obtain basic equipment information and equipment status data, input the multi-objective blocked flow shop scheduling model, solve and obtain the optimal production-maintenance integrated scheduling solution, and schedule the blocked flow shop; Among them, the production-maintenance integrated scheduling scheme includes a production sequence and a maintenance sequence, and the basic equipment information includes the time consumed by each machine to process different workpieces, as well as the processing threshold and maintenance time of each machine; the multi-objective blocking flow shop scheduling model aims to minimize the completion time and total cost, and is solved in turn by diverse search and enhanced search; the diverse search and the enhanced search are production-maintenance iterative optimization modes based on variable neighborhood search of processing sequence granularity and variable neighborhood search of processing fragment granularity, respectively, and the production-maintenance iterative optimization mode includes a production scheduling optimization-maintenance activity following module and a maintenance plan optimization-production activity following module. In the production scheduling optimization-maintenance activity following module, the following steps are repeated multiple times: a variable neighborhood search is performed through a local search operator to obtain a processing sequence, and a corresponding maintenance sequence is generated according to the principle of minimum reliability of the machine; in the maintenance plan optimization-production activity following module, the following steps are repeated multiple times: the maintenance sequence generated by the production scheduling optimization-maintenance activity following module is adjusted according to the forward adjustment condition, and the processing sequence generated by the production scheduling optimization-maintenance activity following module is adjusted.

2. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The local search operator is any one of swap, insert and reverse.

3. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The principle of minimum machine reliability is that before processing a workpiece, the cumulative processing time of the machine processing the workpiece after completing the processing task of the workpiece is calculated. If it exceeds the processing threshold of the machine, the machine needs to be maintained before processing the workpiece; after the maintenance operation is completed, the machine continues to process.

4. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The specific process of adjusting the maintenance sequence generated by the production scheduling optimization-maintenance activity following module according to the forward adjustment conditions is: if moving a maintenance operation forward does not increase the total number of maintenance operations, and the previous idle time of the corresponding machine is sufficient to arrange the maintenance activity, then the maintenance operation and all production and maintenance operations after the maintenance operation will be moved forward as a whole.

5. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The specific process of solving the multi-objective blocking flow shop scheduling model and obtaining the optimal production-maintenance integrated scheduling solution is as follows: S1, randomly generate a processing sequence and set the number of iterations to 1; S2, performing variable neighborhood search on the complete processing sequence through a local search operator to obtain a first processing sequence, and generating a corresponding first maintenance sequence according to the principle of minimum reliability of the machine; S3, adjusting the first maintenance sequence according to the forward adjustment condition, and then adjusting the first processing sequence, and judging whether the first termination condition is met, if not, returning to step S2, if yes, proceeding to step S4; S4. Obtain multiple processing segments according to the first maintenance sequence and the first processing sequence, perform variable neighborhood search on the processing segments through a local search operator to obtain a second processing sequence, and generate a corresponding second maintenance sequence according to a principle of minimum reliability of the machine; S5, adjusting the second maintenance sequence according to the forward adjustment condition, and then adjusting the second processing sequence, and judging whether the second termination condition is met, if not, returning to step S4, if yes, proceeding to step S6; S6. Generate a production-maintenance integrated scheduling plan based on the second processing sequence and the second maintenance sequence, and determine whether the iteration termination condition is met. If not, add 1 to the number of iterations and return to step S2; if so, output the current production-maintenance integrated scheduling plan, which is the optimal production-maintenance integrated scheduling plan.

6. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 5, characterized in that: The first termination condition and the second termination condition are both reaching a preset number of executions, and the iteration termination condition is reaching a preset number of iterations.

7. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 5, characterized in that: According to the first maintenance sequence and the first processing sequence, a specific process of obtaining a plurality of processing segments is as follows: implement in, represents the integrated production-maintenance scheduling solution, ⊙ represents the first maintenance sequence Insert 0 at the corresponding position of the first processing sequence λ; Taking 0 as the node, the production-maintenance integrated scheduling solution Divide into multiple processing segments.

8. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The production scheduling constraints include: The workpieces to be processed must be processed in the same order by all machines; Workpieces that have completed the current operation must remain on the current machine until the next machine is free; A machine can only process one workpiece at a time; A workpiece can only be processed on one machine at a time, and no interruptions are allowed; All machines are considered for maintenance to ensure a minimum reliability level, i.e. maintenance activities are performed on the corresponding machines before the accumulated processing time of each machine exceeds a threshold.

9. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The specific expression of the production scheduling constraint is as follows: D k,1 ≥D (k-1),1 +p k,1 ,k=2,…,n Where n and m represent the number of workpieces and machines respectively, i and z represent the corresponding indexes respectively; k represents the position of the workpiece in the processing sequence; X i,k is a binary variable, when workpiece J i When it is at the kth position in the processing sequence, its value is 1, otherwise it is 0; k,z is a binary variable. When machine z is maintained after the kth workpiece is processed, its value is 1, otherwise it is 0. max Indicates completion time; S k,z With D k,z They represent the start and departure time of the kth workpiece on machine z; r z represents the time required for machine z to undergo maintenance, p i,z represents the processing time of workpiece i on machine z; d z represents the processing threshold of machine z, G k,z represents the cumulative processing time of machine z to process the kth workpiece. When the machine is maintained, G k,z Reset; E k,z It represents the end time of the maintenance activity after processing the kth workpiece on machine z.

10. The method for scheduling a blocked flow shop based on two-layer production-maintenance iterative optimization according to claim 1, characterized in that: The total cost includes maintenance costs and machine operating costs.

Images