Mixed-model assembly line balance sorting optimization method and device

Abstract

The present invention provides a method and device for optimizing the balanced sorting of a mixed-flow assembly line, comprising: obtaining a plurality of balanced sorting schemes for mixed-flow assembly lines, wherein the balanced sorting scheme for the mixed-flow assembly line comprises an E-dimensional mixed-flow assembly line vector, wherein the first-dimensional vector represents a normal working condition, and the second to E-dimensional vectors represent different preventive maintenance conditions, and E is a positive integer, and each-dimensional vector of each of the balanced sorting schemes for the mixed-flow assembly line comprises process allocation information and product production sorting information; optimizing the plurality of balanced sorting schemes for the mixed-flow assembly lines using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting result under each production condition, wherein the multi-objective migratory bird algorithm comprises population crossover operator, balance operator and sorting operator processing.

Description

Technical Field

[0001] The present invention relates to the field of industrial technology, and in particular to a method and device for optimizing the balance sequencing of a mixed-flow assembly line. Background Art

[0002] In order to meet the growing demand for customer customization and the production trend of multiple varieties and small batches, mixed-model assembly lines have become an indispensable production method for manufacturing companies. Unlike traditional single-stream assembly lines that can only assemble one product, mixed-model assembly lines can assemble several products of the same series at the same time. Therefore, mixed-model assembly lines can meet the needs of different customers at the same time. However, since the stations on the mixed-model assembly line are serially connected and there is no cache property between the stations, equipment maintenance at one station is likely to cause production interruption of the entire mixed-model assembly line. In actual production, equipment maintenance is often required at regular intervals, and the entire assembly line will be shut down for equipment preventive maintenance, which also causes huge economic losses. It can be seen from this that separately optimizing the process allocation, product sequencing and equipment preventive maintenance of mixed-model assembly lines cannot guarantee the maintainability of equipment under continuous mixed-model assembly production and production continuity during preventive maintenance.

[0003] Therefore, how to reduce the impact of maintenance on other assembly stations in a mixed-flow assembly line has become an urgent problem to be solved in the industry. Summary of the invention

[0004] The present invention provides a method and device for optimizing the balance sequencing of a mixed-flow assembly line, which are used to solve the defect in the prior art that when a certain station in the mixed-flow assembly line is maintained, other assembly stations will be affected.

[0005] The present invention provides a mixed-flow assembly line balance sequencing optimization method, comprising:

[0006] Acquire multiple mixed-flow assembly line balancing and sorting schemes, wherein the mixed-flow assembly line balancing and sorting schemes include E-dimensional mixed-flow assembly line vectors, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balancing and sorting scheme includes process allocation information and product production sorting information;

[0007] The balanced sorting schemes of the multiple mixed-flow assembly lines are optimized using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting result under each production situation, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0008] According to a mixed-flow assembly line balance sorting optimization method provided by the present invention, after obtaining a plurality of mixed-flow assembly line balance sorting schemes, the method further includes:

[0009] S21, in each case, based on the workstation information, maintenance workstation information and product process processing time information of each of the mixed-model assembly line balanced sequencing schemes, calculating the initial beat of the mixed-model assembly line balanced sequencing scheme under the current production situation;

[0010] S22, constructing a candidate set of optional processes based on the processes that have been directly assigned in the unassigned processes;

[0011] S23, select the target process with the largest weight in the candidate set of optional processes, and assign it to the target station. If the load of any product in the target station is greater than the initial beat value, assign the target process to the next station; otherwise, assign the target process to the current station. If all processes are assigned, proceed to the next step; otherwise, return to step S22

[0012] S24, if the required number of workstations is greater than the given number of workstations, increase the initial beat in the current situation by 1, and return to step S22;

[0013] S25, after all the processes are allocated, the process allocation and product sorting results of each mixed-flow assembly line balance sorting scheme are obtained;

[0014] S26, calculating the completion time and process adjustment number of each mixed-model assembly line balance sequencing solution in each situation based on the process allocation and product sequencing results.

[0015] According to a mixed-flow assembly line balance sorting optimization method provided by the present invention, a multi-objective migratory bird algorithm is performed on the multiple mixed-flow assembly line balance sorting schemes, including:

[0016] S31, determining the leading bird based on the completion time and process adjustment number of each mixed-flow assembly line balanced sequencing scheme in each situation, and randomly placing other mixed-flow assembly line balanced sequencing schemes in the left queue and the right queue of the V-shaped queue to obtain an initial bird flock;

[0017] S32, performing a neighborhood search based on the leader bird to obtain q neighborhood solutions;

[0018] S33, when the performance of the best neighborhood solution among the q neighborhood solutions is better than that of the leader bird, replace the leader bird;

[0019] S34, sharing x neighborhood solutions whose performance values ​​exceed a first preset threshold among the q neighborhood solutions with the birds in the left queue and the birds in the right queue.

[0020] S35, perform neighborhood search on the birds in the left queue respectively, first inherit the x neighborhood solutions inherited by the previous bird, then perform neighborhood search on the bird, and obtain qx neighborhood solutions. If the performance of the best neighborhood solution among the x+(qx) neighborhood solutions is better than that of the bird, replace the bird, and share the x neighborhood solutions whose performance values ​​exceed the first preset threshold with the next bird.

[0021] S36, applying the optimization method in S35 to optimize the flying birds in the right queue

[0022] S37, after repeating steps S32-S36 for h times, randomly select the first flying bird in the left or right queue to replace the leading bird, move the positions of all the flying birds in the queue forward, and place the original leading bird in the last position of the queue.

[0023] According to a mixed-flow assembly line balance sorting optimization method provided by the present invention, the multi-objective migratory bird algorithm includes a balance operator and a sorting operator;

[0024] The balancing operator is used to optimize the process allocation information, and the sorting operator is used to optimize the product production sequence information.

[0025] According to a mixed-model assembly line balance sorting optimization method provided by the present invention, the balance operator includes: a first balance sub-operator, a second balance sub-operator, a third balance sub-operator and a fourth balance sub-operator;

[0026] The first balancing sub-operator is used to randomly select a first random process from the workstation with the largest load, and randomly assign the first random process to a workstation that does not require maintenance and can be assigned;

[0027] The second balancing sub-operator is used to randomly select a second random process from other workstations, and assign the second random process to the workstation with the smallest load. If the second random process cannot be assigned to the workstation with the smallest load, a new process is selected;

[0028] The third balancing sub-operator is used to randomly allocate the processes that are inconsistent with the workstations allocated under the normal working condition and the preventive maintenance condition to other workstations under the normal working condition;

[0029] The fourth balancing sub-operator is used to randomly allocate the processes to which the workstations allocated in the preventive maintenance situation and the normal working situation are inconsistent to other workstations that do not require maintenance under the preventive maintenance situation.

[0030] According to a mixed-model assembly line sorting optimization method provided by the present invention, the sorting operator includes: a first sorting sub-operator and a second sorting sub-operator;

[0031] Wherein, the first sorting sub-operator is used to insert the model with the smallest total processing time into a random position before the model;

[0032] The second sorting sub-operator is used to exchange the model with the smallest total processing time with the model with the largest total processing time that is arranged in front of the model.

[0033] The present invention also provides a mixed-flow assembly line balance sequencing optimization device, comprising:

[0034] An acquisition module is used to acquire multiple mixed-flow assembly line balance sorting schemes, wherein the mixed-flow assembly line balance sorting scheme includes an E-dimensional mixed-flow assembly line vector, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balance sorting scheme includes process allocation information and product production sorting information;

[0035] The optimization module is used to optimize the balance and sorting schemes of the multiple mixed-flow assembly lines using a multi-objective migratory bird algorithm to obtain an optimal balance and sorting plan, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0036] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned mixed-flow assembly line balance sorting optimization methods is implemented.

[0037] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for optimizing the balanced sorting of a mixed-flow assembly line as described in any one of the above is implemented.

[0038] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned mixed-flow assembly line balance sorting optimization methods.

[0039] The present invention provides a mixed-flow assembly line balance sorting optimization method and device, which fully considers the preventive maintenance plan and the mixed-flow assembly line balance sorting problem, adjusts the process and product production sequence of the mixed-flow assembly line under various production situations according to a given target maintenance situation, and then quickly obtains a high-quality solution through an improved multi-objective migratory bird optimization algorithm, and finally obtains a sorting result with less impact on other assembly lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flow chart of a method for optimizing the balance and sorting of a mixed-flow assembly line provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of the field structure provided for the embodiments of the present application;

[0043] Figure 3 An example diagram of a crossover operator provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of a mixed-flow assembly line balance sorting optimization device provided in an embodiment of the present application;

[0045] Figure 5 An example physical structure diagram of an electronic device is provided. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Specifically, in the embodiment of the present application, a multi-objective mixed integer programming model for balancing and sequencing mixed-flow assembly lines considering preventive maintenance is established. Table 1 is a description of the main parameters in the model, as shown in Table 1 below:

[0048] Table 1 Description of main parameters

[0049]

[0050] Table 2 is a description of the main variables in the model, as shown in Table 2:

[0051] Table 2 Description of main variables

[0052]

[0053] The multi-objective mixed integer programming model for balancing and sequencing mixed-flow assembly lines considering preventive maintenance described in the embodiments of the present application is specifically:

[0054] Objective function:

[0055]

[0056]

[0057]

[0058] Constraints:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] Among them, formula (1) indicates that the objective function is to minimize the maximum completion time makespan of normal working conditions. Formula (2) indicates that the objective function is to minimize the maximum makespan in each maintenance situation. Formula (3) indicates that the objective function is to minimize the maximum process change of all maintenance situations. Formula (4) indicates that each process can only be assigned to a workstation that does not require maintenance in each situation. Formula (5) indicates that each workstation that does not require maintenance should be assigned at least one process. Formula (6) indicates that a process cannot be assigned to a workstation that requires maintenance. Formula (7) indicates the precedence constraint, that is, a process can only be assigned after all its predecessors are completed. Formula (8) indicates that all products must participate in the sorting. Formula (9) indicates that the number of products participating in the sorting is equal to the number of products in the master production plan. Formula (10) calculates the total processing time of product m at workstation k under situation e. Formula (11) calculates the completion time of the first product at the first workstation. Formula (12) restricts that the next workstation can only start processing after the processing of the previous workstation is completed. Formula (13) ensures that the products in the rear can only be processed after the products in the front are completed. Formulas (14) and (15) calculate the process change of process i of product m between normal situation and maintenance situation.

[0072] Figure 1 The flow chart of the mixed-flow assembly line balance sorting optimization method provided in the embodiment of the present application is as follows: Figure 1 As shown, including:

[0073] Step 110, obtaining a plurality of mixed-flow assembly line balancing and sequencing schemes, wherein the mixed-flow assembly line balancing and sequencing schemes include E-dimensional mixed-flow assembly line vectors, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balancing and sequencing scheme includes process allocation information and product production sequencing information;

[0074] Specifically, the balanced sorting scheme for the mixed-flow assembly line described in the embodiments of the present application may include a balanced sorting scheme for normal working conditions, and may also include a balanced sorting scheme for different preventive maintenance conditions. More specifically, the first-dimensional vector represents a normal working condition, and the 2nd to E-dimensional vectors represent different preventive maintenance conditions.

[0075] Each mixed-flow assembly line balanced sorting scheme described in the embodiments of the present application includes two fragments, namely, process allocation information and product production sequence information, wherein the process allocation information determines the allocation of processes and contains N elements, and the product production sequence information determines the production sequence of products and contains M elements. Therefore, each mixed-flow assembly line balanced sorting scheme includes a total of E×(N+M) elements. Each element is a random number between 0 and 1, representing the weight of the corresponding process or product. For the process allocation information, if a process obtains the maximum weight and its direct predecessor has been processed, then this process can be processed. For the product production sequence information, the product with the maximum weight will be processed first, and other products will be processed after it.

[0076] In the initialization phase of the algorithm population, each individual will randomly generate E×(N+M) elements, of which the first N+M elements belong to normal working conditions, and N+M+1 to (N+M)×2 belong to the first preventive maintenance condition.

[0077] Specifically, the preventive maintenance scenario described in the embodiments of the present application specifically refers to a scenario in which a certain workstation of a mixed-flow assembly line that is pre-planned needs to be maintained.

[0078] In the embodiment of the present application, the influence of the maintenance station under maintenance is fully considered, and the weights of the process and product under each situation are generated on the basis of satisfying the process priority relationship. According to the setting of the initial beat, the process is allocated to each station to balance the load of each station.

[0079] Step 120, using a multi-objective migratory bird algorithm to optimize the multiple mixed-flow assembly line balance sorting schemes to obtain the optimal process allocation plan and product sorting results under each production situation, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0080] Specifically, the multi-objective migratory bird algorithm described in the embodiments of the present application designs a random key encoding and decoding method based on the priority relationship between processes, and also designs a neighborhood structure based on specific problems.

[0081] At the same time, the embodiment of the present application also designs 6 restart mechanisms based on specific problems to prevent the algorithm from falling into local optimality, and proposes an inter-population crossover operator to improve communication within the population.

[0082] In an embodiment of the present application, by fully considering the preventive maintenance plan and the balanced sequencing problem of the mixed-flow assembly line, according to the given target maintenance situation, the process and product production order of the mixed-flow assembly line under various production situations are adjusted to obtain multiple balanced sequencing schemes for mixed-flow assembly lines, and then optimized through the improved multi-objective migratory bird optimization algorithm to quickly obtain a high-quality solution, and finally obtain a balanced sequencing scheme with less impact on other assembly stations.

[0083] More specifically, S21, in each case, based on the workstation information, maintenance workstation information and product process processing time information of each of the mixed-model assembly line balanced sequencing schemes, calculating the initial beat of the mixed-model assembly line balanced sequencing scheme under the current production situation;

[0084] S22, constructing a candidate set of optional processes based on the processes that have been directly assigned in the unassigned processes;

[0085] S23, select the target process with the largest weight in the candidate set of optional processes, and assign it to the target station. If the load of any product in the target station is greater than the initial beat value, assign the target process to the next station; otherwise, assign the target process to the current station. If all processes are assigned, proceed to the next step; otherwise, return to step S22

[0086] S24, if the required number of workstations is greater than the given number of workstations, increase the initial beat in the current situation by 1, and return to step S22;

[0087] S25, after all the processes are allocated, the process allocation and product sorting results of each mixed-flow assembly line balance sorting scheme are obtained;

[0088] S26, calculating the completion time and process adjustment number of each mixed-model assembly line balance sequencing solution in each situation based on the process allocation and product sequencing results.

[0089] Specifically, each situation described in the embodiments of the present application includes various preventive maintenance situations and normal working situations.

[0090] Specifically, the tact time described in the embodiments of the present application refers to the maximum load on all workstations after the process is distributed to all workstations.

[0091] In the embodiment of the present application, the initial beat is calculated according to the initial beat calculation formula, specifically:

[0092]

[0093] Based on the assigned processes of the immediate predecessors among the unassigned processes, a candidate set of optional processes is constructed;

[0094] Select all the immediately preceding assigned processes from the unassigned processes and add them to the candidate set; select the process with the largest weight from the candidate set. If there is no process in the current candidate set, return to step 22 and rebuild the candidate set of optional processes;

[0095] Assign the process to the current workstation. If the load of any product at the target workstation is greater than the initial beat value, assign the target process to the next workstation; otherwise, assign the target process to the current workstation. If all process assignments are completed, proceed to the next step.

[0096] If the required number of workstations is greater than the given number of workstations, the initial beat in the current situation is increased by 1, and the candidate set of optional processes is reconstructed. After all processes are allocated, the process allocation and product sorting results of each mixed-flow assembly line balanced sorting scheme are obtained.

[0097] The above steps allocate the processes to each workstation on the basis of satisfying the priority relationship. For the product production sequence information, the product sorting result can be obtained by sorting according to the weight of the corresponding fragment elements, and the completion time and process adjustment number of each mixed-flow assembly line balanced sorting scheme in each case are calculated based on the process allocation and product sorting results.

[0098] In the embodiment of the present application, by fully considering the workstation information that needs to be maintained under the target maintenance situation, the processes of the mixed-flow assembly line are reallocated, thereby ensuring the rationality of the subsequent allocation.

[0099] Optionally, performing a multi-objective migratory bird algorithm optimization on the multiple mixed-flow assembly line balance sorting schemes includes:

[0100] S31, determining the leading bird based on the completion time and process adjustment number of each mixed-flow assembly line balanced sequencing scheme in each situation, and randomly placing other mixed-flow assembly line balanced sequencing schemes in the left queue and the right queue of the V-shaped queue to obtain an initial bird flock;

[0101] S32, performing a neighborhood search based on the leader bird to obtain q neighborhood solutions;

[0102] S33, when the performance of the best neighborhood solution among the q neighborhood solutions is better than that of the leader bird, replace the leader bird;

[0103] S34, sharing x neighborhood solutions whose performance values ​​exceed a first preset threshold among the q neighborhood solutions with the birds in the left queue and the birds in the right queue.

[0104] S35, perform neighborhood search on the birds in the left queue respectively, first inherit the x neighborhood solutions inherited by the previous bird, then perform neighborhood search on the bird, and obtain qx neighborhood solutions. If the performance of the best neighborhood solution among the x+(qx) neighborhood solutions is better than that of the bird, replace the bird, and share the x neighborhood solutions whose performance values ​​exceed the first preset threshold with the next bird.

[0105] S36, applying the optimization method in S35 to optimize the flying birds in the right queue

[0106] S37, after repeating steps S32-S36 for h times, randomly select the first flying bird in the left or right queue to replace the leading bird, move the positions of all the flying birds in the queue forward, and place the original leading bird in the last position of the queue.

[0107] Specifically, the embodiment of the present application calculates the makespan and process adjustment number of each individual under each situation based on the process allocation and product sorting results obtained in the above embodiment, that is, the mixed-flow assembly line balanced sorting scheme. The non-dominated solutions are stored in an external archive set, and the mixed-flow assembly line balanced sorting scheme with the largest crowding distance is selected as the leading bird, and the other birds are randomly placed on the left and right sides of the V-shaped queue, thereby completing the construction of the initial bird flock.

[0108] The multi-objective migratory bird algorithm described in the embodiment of the present application has a specific domain structure. Since the mixed-flow assembly line balancing and sorting problem considering preventive maintenance situations includes E production situations and two sub-problems, it is necessary to update the fragments of the two sub-problems in E situations. When the neighborhood structure generates a new solution, it randomly selects the weights of two elements in the two sub-problems of balance and sorting in each situation and exchanges them. Figure 2 A schematic diagram of the field structure provided in the embodiment of the present application, such as Figure 2 As shown in Figure 1, it shows the update of the neighborhood structure to the balancing and sorting fragments in a certain case. This case contains 11 processes and 3 products. The neighborhood structure randomly selects the 3rd and 6th elements of the balancing fragment and the 1st and 3rd elements of the sorting stage and crosses them.

[0109] When the neighborhood structure is applied to generate q neighborhood solutions of the leading bird, if the best neighborhood solution is better than the leading bird, the leading bird will be updated. The x unused better neighborhood solutions are shared with the birds on the left and right. When the neighborhood structure is used to update the birds on the left and right, first accept the x neighborhood solutions passed by the previous bird and generate qx neighborhood solutions according to the neighborhood structure. When the best individual among these neighborhood solutions is better than the current bird, the bird is replaced with the best individual. Then the q unused better neighborhood solutions are passed to the birds below.

[0110] More specifically, for the first bird on the left, accept the x neighborhood solutions shared by the leading bird, and apply the neighborhood structure to generate qx neighborhood solutions. Evaluate the performance of these q neighborhood solutions. If the performance of the optimal neighborhood solution is better than that of the bird, replace the bird. Otherwise, do not replace it. Share the x unused neighborhood solutions with better performance with the birds below it. Update the remaining birds on the left one by one

[0111] For the first bird on the right, accept the x neighborhood solutions shared by the leading bird and apply the neighborhood structure to generate qx neighborhood solutions. Evaluate the performance of the q neighborhood solutions. If the performance of the best neighborhood solution is better than that of the bird, replace the bird. Otherwise, do not replace it. Share the x unused neighborhood solutions with the birds below it.

[0112] At the same time, the multi-target migratory bird algorithm described in the embodiment of the present application also has a crossover operator within the population. Figure 3 The crossover operator example diagram provided in the embodiment of the present application is as follows: Figure 3 As shown in the figure, for each bird in the population, the operator randomly finds another bird from the population, and then updates the balance and sorting segments in each case. First, two sub-segments in the corresponding segment are randomly selected and then exchanged. Once the newly generated solution is better than the parent solution, the parent solution will be replaced.

[0113] In the embodiment of the present application, a better sorting result can be effectively determined by a multi-objective flying bird algorithm.

[0114] Optionally, the multi-objective migratory bird algorithm includes a balancing operator and a sorting operator;

[0115] The balancing operator is used to optimize the process allocation information, and the sorting operator is used to optimize the product production sequence information.

[0116] The balancing operators include: a first balancing sub-operator, a second balancing sub-operator, a third balancing sub-operator and a fourth balancing sub-operator;

[0117] The first balancing sub-operator is used to randomly select a first random process from the workstation with the largest load, and randomly assign the first random process to a workstation that does not require maintenance and can be assigned;

[0118] The second balancing sub-operator is used to randomly select a second random process from other workstations, and assign the second random process to the workstation with the smallest load. If the second random process cannot be assigned to the workstation with the smallest load, a new process is selected;

[0119] The third balancing sub-operator is used to randomly allocate the processes that are inconsistent with the workstations allocated under the normal working situation and the target maintenance situation to other workstations under the normal working situation;

[0120] The fourth balancing sub-operator is used to randomly allocate the processes to which the workstations allocated under the normal working situation and the target maintenance situation are inconsistent to other workstations that do not require maintenance under the target maintenance situation.

[0121] The sorting operator includes: a first sorting sub-operator and a second sorting sub-operator;

[0122] Wherein, the first sorting sub-operator is used to insert the model with the smallest total processing time into a random position before the model;

[0123] The second sorting sub-operator is used to exchange the model with the smallest total processing time with the model with the largest total processing time that is arranged in front of the model.

[0124] Specifically, since the basic migratory bird algorithm belongs to a neighborhood search algorithm, it is easy to fall into a local optimum. Therefore, this application designs two restart operators that take into account the characteristics of this problem, 4 balancing sub-operators and 2 sorting sub-operators, to optimize the balancing segment and the sorting segment respectively.

[0125] The balancing operator mainly generates new individuals by adjusting the processing order of the process. This type of operator mainly includes four operators: BO_1, BO_2, BO_3, and BO_4.

[0126] BO_1: Randomly select a process from the workstation with the largest load and randomly assign it to a workstation that does not require maintenance and can be assigned.

[0127] BO_2: Randomly select a process from other stations and assign it to the station with the smallest load. If the process cannot be assigned to the station, select a new process.

[0128] BO_3: For normal working conditions, find the processes whose assigned workstations are inconsistent between this condition and the maintenance condition, and randomly assign them to other workstations.

[0129] BO_4: For maintenance situations, find the workstations that are inconsistent with those in normal situations and have the largest process adjustment, and randomly assign them to other workstations that do not require maintenance.

[0130] The sorting operator mainly generates new individuals by adjusting the model's production order. It mainly includes two operators: SO_1 and SO_2.

[0131] SO_1: Insert the model with the minimum total processing time at a random position before it.

[0132] SO_2: Exchange the model with the smallest total processing time with the model with the largest time in front of it.

[0133] The balancing operator and sorting operator in the embodiment of the present application can effectively prevent the multi-objective migratory bird algorithm from falling into local optimality and ensure the reliability of the final result.

[0134] Optionally, in the embodiment of the present application, the processing time of each product is generated by disturbing the time in the benchmark case. Several maintenance situations are randomly generated for each case, and the workstations that need maintenance are randomly generated. The information of these cases is shown in Table 3. The first and second columns show the serial number of the case and the number of processes. The third and fourth columns show the number of workstations and the number of products in the corresponding case. The last column shows the serial number of the workstation that needs maintenance in each case.

[0135] Table 3. Test case information

[0136]

[0137]

[0138]

[0139] All algorithms are programmed in Matlab. This application uses two evaluation indicators, generation distance and hypervolume ratio, to evaluate the convergence and distribution of the proposed algorithm. The smaller the generation distance, the closer the obtained non-dominated solution set is to the true frontier solution. The larger the hypervolume ratio value, the better the distribution of the obtained non-dominated solution in the solution space.

[0140] This application compares the proposed IMOMBO with six multi-objective optimization techniques to verify the superiority of the proposed method. The comparison algorithms mainly include MOALO, MODA, MOWOA, MOPSO and MOMFO. After testing 51 cases, the final comparison results are shown in Table 4-5.

[0141] From Table 4, it can be found that the average generation distances of MOALO, MODA, MOWOA, MOPSO and MOMFO algorithms are 0.085, 0.077, 0.069, 0.058 and 0.062, respectively. The average generation distance of the proposed algorithm is 0.006, which is significantly smaller than other algorithms, indicating that the proposed algorithm is better than the compared multi-objective algorithms in terms of convergence. In Table 5, the average hypervolume ratios of MOALO, MODA, MOWOA, MOPSO, MOMFO and IMOMBO algorithms are 0.608, 0.611, 0.653, 0.699, 0.693 and 1.151, respectively. This shows that the proposed algorithm is also significantly better than other algorithms in terms of distribution. This shows that the proposed algorithm has good convergence and distribution in solving the mixed-flow assembly line balance sorting problem considering preventive maintenance.

[0142] Table 4. Average generation distance of all algorithms

[0143]

[0144]

[0145]

[0146]

[0147] Table 5. Average HVR values ​​of all algorithms

[0148]

[0149]

[0150]

[0151] The mixed-flow assembly line balance sequencing optimization device provided by the present invention is described below. The mixed-flow assembly line balance sequencing optimization device described below and the mixed-flow assembly line balance sequencing optimization method described above can be referred to each other.

[0152] Figure 4 The schematic diagram of the structure of the mixed-flow assembly line balance sorting optimization device provided in the embodiment of the present application is as follows: Figure 4 As shown, including:

[0153] The acquisition module 410 is used to acquire multiple mixed-flow assembly line balance sequencing schemes, wherein the mixed-flow assembly line balance sequencing schemes include E-dimensional mixed-flow assembly line vectors, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balance sequencing scheme includes process allocation information and product production sequencing information;

[0154] The adjustment module 420 is used to perform multi-objective migratory bird algorithm optimization on the multiple mixed-flow assembly line balance sorting schemes to obtain an optimal balance and sorting plan, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0155] The embodiment of the present invention fully considers the preventive maintenance plan and the balanced sorting problem of mixed-flow assembly lines, adjusts the process allocation and product production sequence of mixed-flow assembly lines under various production situations according to given target maintenance situations, and obtains multiple balanced sorting schemes for mixed-flow assembly lines. Then, the schemes are optimized through an improved multi-objective migratory bird optimization algorithm to quickly obtain solutions with higher quality, and finally obtain a sorting result with less impact on other assembly lines.

[0156] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute a mixed-flow assembly line balance sorting optimization method, the method comprising: obtaining multiple mixed-flow assembly line balance sorting schemes, wherein the mixed-flow assembly line balance sorting schemes include E-dimensional mixed-flow assembly line vectors, the first-dimensional vector represents a normal working situation, the second to E-dimensional vectors represent different preventive maintenance situations, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balance sorting scheme includes process allocation information and product production sorting information; the multiple mixed-flow assembly line balance sorting schemes are optimized using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting result under each production situation, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0157] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, 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, and the computer software product is stored in a storage medium, including a number of 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, etc. Various media that can store program codes.

[0158] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the mixed-flow assembly line balance sorting optimization method provided by the above-mentioned methods, the method including: obtaining multiple mixed-flow assembly line balance sorting schemes, wherein the mixed-flow assembly line balance sorting scheme includes an E-dimensional mixed-flow assembly line vector, the first-dimensional vector represents a normal working condition, and the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each of the mixed-flow assembly line balance sorting schemes includes process allocation information and product production sorting information; the multiple mixed-flow assembly line balance sorting schemes are optimized using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting results under each production condition, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing.

[0159] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the mixed-flow assembly line balance sorting optimization method provided by the above-mentioned methods, the method comprising: obtaining multiple mixed-flow assembly line balance sorting schemes, wherein the mixed-flow assembly line balance sorting scheme comprises an E-dimensional mixed-flow assembly line vector, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each-dimensional vector of each of the mixed-flow assembly line balance sorting schemes comprises process allocation information and product production sorting information; optimizing the multiple mixed-flow assembly line balance sorting schemes using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting results under each production condition, wherein the multi-objective migratory bird algorithm comprises population crossover operator, balance operator and sorting operator processing.

[0160] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0161] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the balance sorting of a mixed-model assembly line, characterized in that: include: Acquire multiple mixed-flow assembly line balancing and sorting schemes, wherein the mixed-flow assembly line balancing and sorting schemes include E-dimensional mixed-flow assembly line vectors, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balancing and sorting scheme includes process allocation information and product production sorting information; The balanced sorting schemes of the multiple mixed-flow assembly lines are optimized using a multi-objective migratory bird algorithm to obtain the optimal process allocation plan and product sorting result under each production situation, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing; After obtaining the balanced sorting schemes for multiple mixed-flow assembly lines, the method further includes: S21, in each case, based on the workstation information, maintenance workstation information and product process processing time information of each of the mixed-model assembly line balanced sequencing schemes, calculating the initial beat of the mixed-model assembly line balanced sequencing scheme under the current production situation; S22, constructing a candidate set of optional processes based on the processes that have been directly assigned in the unassigned processes; S23, selecting the target process with the largest weight in the candidate set of optional processes, and assigning it to the target station. If the load of any product in the target station is greater than the initial beat value, the target process is assigned to the next station; otherwise, the target process is assigned to the current station. If all processes are assigned, proceed to the next step; otherwise, return to step S22; S24, if the required number of workstations is greater than the given number of workstations, increase the initial beat in the current situation by 1, and return to step S22; S25, after all the processes are allocated, the process allocation and product sorting results of each mixed-flow assembly line balance sorting scheme are obtained; S26, calculating the completion time and process adjustment number of each mixed-model assembly line balance sequencing solution in each case based on the process allocation and product sequencing results; The method of optimizing the balanced sorting schemes of the multiple mixed-flow assembly lines using a multi-objective migratory bird algorithm includes: S31, determining the leading bird based on the completion time and process adjustment number of each mixed-flow assembly line balanced sequencing scheme in each situation, and randomly placing other mixed-flow assembly line balanced sequencing schemes in the left queue and the right queue of the V-shaped queue to obtain an initial bird flock; S32, performing a neighborhood search based on the leader bird to obtain q neighborhood solutions; S33, when the performance of the best neighborhood solution among the q neighborhood solutions is better than that of the leader bird, replace the leader bird; S34, sharing x neighborhood solutions whose performance values ​​exceed a first preset threshold among the q neighborhood solutions with the birds in the left queue and the birds in the right queue; S35, performing neighborhood search on the birds in the left queue in turn, first inheriting the x neighborhood solutions inherited by the previous bird, then performing neighborhood search on the bird to obtain qx neighborhood solutions, and replacing the bird when the performance of the best neighborhood solution among x+(qx) neighborhood solutions is better than that of the bird, and sharing the x neighborhood solutions whose performance values ​​exceed a first preset threshold among the neighborhood solutions with the next bird; S36, applying the optimization method in S35 to optimize the flying birds in the right queue; S37, after repeating steps S32-S36 for h times, randomly select the first flying bird in the left or right queue to replace the leading bird, move the positions of all the flying birds in the queue forward, and place the original leading bird in the last position of the queue.

2. The method for optimizing the balance and sorting of mixed-model assembly lines according to claim 1, characterized in that: The multi-objective migratory bird algorithm includes a balance operator and a sorting operator; The balancing operator is used to optimize the process allocation information, and the sorting operator is used to optimize the product production sequence information.

3. The method for optimizing the balance sorting of mixed-model assembly lines according to claim 2, characterized in that: The balancing operators include: a first balancing sub-operator, a second balancing sub-operator, a third balancing sub-operator and a fourth balancing sub-operator; The first balancing sub-operator is used to randomly select a first random process from the workstation with the largest load, and randomly assign the first random process to a workstation that does not require maintenance and can be assigned; The second balancing sub-operator is used to randomly select a second random process from other workstations, and assign the second random process to the workstation with the smallest load. If the second random process cannot be assigned to the workstation with the smallest load, a new process is selected; The third balancing sub-operator is used to randomly allocate the processes that are inconsistent with the workstations allocated under the normal working condition and the preventive maintenance condition to other workstations under the normal working condition; The fourth balancing sub-operator is used to randomly allocate the processes to which the workstations allocated in the preventive maintenance situation and the normal working situation are inconsistent to other workstations that do not require maintenance.

4. The method for optimizing the balance and sorting of mixed-model assembly lines according to claim 2, characterized in that: The sorting operator includes: a first sorting sub-operator and a second sorting sub-operator; Wherein, the first sorting sub-operator is used to insert the model with the smallest total processing time into a random position before the model; The second sorting sub-operator is used to exchange the model with the smallest total processing time with the model with the largest total processing time that is arranged in front of the model.

5. A device for optimizing the balance and sorting of mixed-model assembly lines, characterized in that: include: An acquisition module is used to acquire multiple mixed-flow assembly line balance sorting schemes, wherein the mixed-flow assembly line balance sorting scheme includes an E-dimensional mixed-flow assembly line vector, the first-dimensional vector represents a normal working condition, the second to E-dimensional vectors represent different preventive maintenance conditions, E is a positive integer, and each dimensional vector of each mixed-flow assembly line balance sorting scheme includes process allocation information and product production sorting information; An optimization module, used for performing a multi-objective migratory bird algorithm optimization on the balance and sorting schemes of the multiple mixed-flow assembly lines to obtain an optimal balance and sorting plan, wherein the multi-objective migratory bird algorithm includes population crossover operator, balance operator and sorting operator processing; Wherein, the device is also used for: S21, in each case, based on the workstation information, maintenance workstation information and product process processing time information of each of the mixed-model assembly line balanced sequencing schemes, calculating the initial beat of the mixed-model assembly line balanced sequencing scheme under the current production situation; S22, constructing a candidate set of optional processes based on the processes that have been directly assigned in the unassigned processes; S23, selecting the target process with the largest weight in the candidate set of optional processes, and assigning it to the target station. If the load of any product in the target station is greater than the initial beat value, the target process is assigned to the next station; otherwise, the target process is assigned to the current station. If all processes are assigned, proceed to the next step; otherwise, return to step S22; S24, if the required number of workstations is greater than the given number of workstations, increase the initial beat in the current situation by 1, and return to step S22; S25, after all the processes are allocated, the process allocation and product sorting results of each mixed-flow assembly line balance sorting scheme are obtained; S26, calculating the completion time and process adjustment number of each mixed-model assembly line balance sequencing solution in each case based on the process allocation and product sequencing results; The balanced sorting schemes of the multiple mixed-flow assembly lines are optimized using a multi-objective migratory bird algorithm, including: S31, determining the leading bird based on the completion time and process adjustment number of each mixed-flow assembly line balanced sequencing scheme in each situation, and randomly placing other mixed-flow assembly line balanced sequencing schemes in the left queue and the right queue of the V-shaped queue to obtain an initial bird flock; S32, performing a neighborhood search based on the leader bird to obtain q neighborhood solutions; S33, when the performance of the best neighborhood solution among the q neighborhood solutions is better than that of the leader bird, replace the leader bird; S34, sharing x neighborhood solutions whose performance values ​​exceed a first preset threshold among the q neighborhood solutions with the birds in the left queue and the birds in the right queue; S35, performing neighborhood search on the birds in the left queue in turn, first inheriting the x neighborhood solutions inherited by the previous bird, then performing neighborhood search on the bird to obtain qx neighborhood solutions, and replacing the bird when the performance of the best neighborhood solution among x+(qx) neighborhood solutions is better than that of the bird, and sharing the x neighborhood solutions whose performance values ​​exceed a first preset threshold among the neighborhood solutions with the next bird; S36, applying the optimization method in S35 to optimize the flying birds in the right queue; S37, after repeating steps S32-S36 for h times, randomly select the first flying bird in the left or right queue to replace the leading bird, move the positions of all the flying birds in the queue forward, and place the original leading bird in the last position of the queue.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the mixed-flow assembly line balance sorting optimization method as described in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the mixed-flow assembly line balance sorting optimization method as claimed in any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the mixed-flow assembly line balance sorting optimization method as claimed in any one of claims 1 to 4 is implemented.

Images