A multi-objective parallel method for optimizing disassembly line of waste products based on worker skills

Abstract

The application discloses a kind of based on worker skill number multi-objective parallel waste product disassembly line optimization method, and relates to a kind of waste product disassembly line optimization method, the product information and disassembly line information of the product to be disassembled and the disassembly skill information of worker are constructed disassembly line balance model;With the minimum worker skill number and the net profit of product disassembly maximization as target, construct objective function;Set the constraint condition of disassembly line balance optimization model;Coding mechanism based on priority relation matrix is coded to disassembly task, generates feasible solution, decoding;Get the task allocation scheme of the product to be disassembled;In actual case, IDBA is compared with NSGA-II, NSGA-III, SPEA-II, ESPEA, MOEA / D and five kinds of algorithms.The application proposes an improved discrete bat algorithm according to the actual case of disassembly line, optimizes multi-objective parallel disassembly line balance problem, and significantly improves task and skill allocation efficiency.

Description

TECHNICAL FIELD

[0001] The application relates to a waste product disassembly line optimization method, in particular to a waste product disassembly line optimization method based on worker skill number multi-objective parallelism. BACKGROUND

[0002] The rapid development of information technology has profoundly changed the daily life of modern society. At the same time, the vigorous development of modern manufacturing industry has led to the decline of product cost and the rise of market demand, which not only speeds up the product update and replacement, but also significantly increases the generation rate of waste products. In the face of the dual pressures of resource waste and environmental pollution, the recycling and reprocessing of waste products are particularly urgent. In this process, the disassembly link plays a crucial role, which involves separating useful parts and potentially harmful substances from waste products. Through meticulous disassembly, these parts can be recycled and utilized, and after strict testing, repair and upgrading, they will re-enter the remanufacturing process, thereby maximizing the value of waste products and minimizing their impact on the environment and human health.

[0003] The disassembly of waste products requires high standardization and specialization of the industry. In order to ensure the efficiency and environmental friendliness of the disassembly process, disassembly enterprises need to develop fine disassembly strategies for various products. In the disassembly operation, not only the balance of the production line needs to be maintained, but also the actual operation of the workers needs to be fully considered, including the number of skills possessed by the workers, the applicability of disassembly tools, the rationality of disassembly sequence, and the priority of disassembly of parts. SUMMARY

[0004] The purpose of the present application is to provide a waste product disassembly line optimization method based on worker skill number multi-objective parallelism, which improves the overall efficiency of the disassembly line by considering the number of skills possessed by the workers, focuses on optimizing the skill allocation of the workers, so that each worker can maximize their potential in the disassembly process, achieve balanced operation of the disassembly line, improve disassembly efficiency, and reduce the risk of operation errors and injuries.

[0005] The present application provides the following technical solutions:

[0006] A waste product disassembly line optimization method based on worker skill number multi-objective parallelism, the method comprising the following processes:

[0007] S1, constructing a disassembly line balance model according to product information and disassembly line information of the products to be disassembled, and disassembly skill information possessed by the workers;

[0008] S2, constructing a target function with the goal of minimizing the number of skills needed to be mastered by the workers and maximizing the net profit of product disassembly;

[0009] S3 sets the constraint condition of the disassembly line balance optimization model;

[0010] S4 encodes the disassembly task based on the coding mechanism of the priority relationship matrix, generates a feasible solution, and decodes;

[0011] S5 solves the multi-objective disassembly line balance optimization model by using an improved discrete bat algorithm IDBA, and obtains a task allocation scheme of the products to be disassembled;

[0012] S6 compares the improved IDBA with NSGA-II, NSGA-III, SPEA-II, ESPEA and MOEA / D in actual cases.

[0013] The information of the products to be disassembled includes the number of disassembly tasks, the number of components, the number of parts, the unit time disassembly cost and the recycling cost of the product components, the priority relationship between the disassembly tasks of the product, the disassembly relationship between the tasks and the components, and the disassembly time of each task on the disassembly line; the disassembly line information includes the cycle period of the disassembly line; all the disassembly skill information of the workers includes the number of disassembly skills mastered by the workers in the process of disassembling the product.

[0014] The multi-objective parallel waste product disassembly line optimization method based on the number of skills of workers, characterized in that the method takes minimizing the number of skills required to be mastered by workers and maximizing the net profit of product disassembly as the target, and constructs a target function including:

[0015] maxf1

[0016] minf2

[0017]

[0018] The multi-objective parallel waste product disassembly line optimization method based on the number of skills of workers, wherein the coding mechanism based on the priority relationship matrix encodes the disassembly task, generates a feasible solution, and decodes, including:

[0019] S51, the priority relationship matrix of the task is obtained according to the and / or graph of the task execution order in the product;

[0020] S52, according to the priority relationship matrix of the task, a vector π is introduced to represent a disassembly task sequence of the product, each element in π corresponds to a number of the task and is arranged in order; this vector π is a task coding method; by introducing the vector π, the system can digitally represent the disassembly sequence; for some complex disassembly tasks, the vector π ensures that each task can follow the rules of the priority relationship matrix, thereby maintaining the logical order of the tasks;

[0021] S53, generating a feasible solution by sequential insertion method; inserting the disassembly task sequence of one product into the task sequence of another product, ensuring that the disassembly tasks are sequentially assigned, meeting the priority and time requirements of the tasks;

[0022] S54, assigning the generated task disassembly sequence to different workstations, ensuring that the task time of the workstation does not exceed its cycle time; in this decoding process, not only the disassembly sequence of the task is considered, but also the constraint condition of the task time is met.

[0023] The waste product disassembly line optimization method based on the number of worker skills and multi-objective parallelism, the improved discrete bat optimization algorithm is used to solve the multi-objective disassembly line balancing optimization model, and the task allocation scheme of the product to be disassembled includes:

[0024] S61, population initialization and adjustment;

[0025] S62, updating individuals by using the cross operation PPX based on priority reservation crossing and the mutation operation based on the direct previous task sequence;

[0026] S63, updating the population by non-dominated sorting, selecting the Pareto optimal solution set, and gradually iterating until the stop condition is met;

[0027] S64, obtaining a disassembly line balancing strategy that meets the balancing constraint condition.

[0028] The waste product disassembly line optimization method based on the number of worker skills and multi-objective parallelism, the population initialization and adjustment in S61 includes:

[0029] S611, each individual in the population represents a feasible disassembly sequence; the population is generated by random insertion method, ensuring the randomness and diversity of each individual; the specific method is to randomly insert the disassembly task sequence of product II into the disassembly task sequence of product I;

[0030] S612, when generating the feasible disassembly sequence of the parallel disassembly line, it is necessary to ensure that the disassembly tasks of each product follow their priority relationship; however, the randomly generated sequence may not necessarily meet these constraints, so further adjustment is necessary;

[0031] S613, the adjustment process is divided into three stages: addition operation, adding the missed tasks to the existing infeasible sequence according to the priority relationship; removal operation, removing the conflicting tasks in the task sequence; exchange operation, after the completion of the previous two stages, if some tasks still do not meet the priority relationship, adjust the task order.

[0032] The one based on the number of worker skills multi-objective parallel waste product disassembly line optimization method, the method for updating the individual by the priority reservation crossing operation PPX and the variation operation based on the direct predecessor task sequence in S62 includes:

[0033] S621, the priority reservation crossing operation, which aims to inherit the excellent genes in the parent when updating the individual, and ensure that the priority relationship of the disassembly task is not damaged;

[0034] S622, the process of the priority reservation crossing operation is as follows: first, input two individuals, respectively, the current best individual π f and the individual π m selected randomly from the current Pareto solution set, and a randomly generated mask sequence MASK; if the mask bit is 1, the corresponding parent individual π m Element is filled into the new individual π new ; if the mask bit is 0, the corresponding parent individual π f Element is filled into the new individual π new ; finally output: the generated new individual π new ;

[0035] S623, the variation operation based on the direct predecessor task sequence, which can adjust the individual locally while maintaining the basic structure of the individual, thereby improving the diversity of the solution and helping the algorithm jump out of the local optimum; in this optimization method, the variation operation is based on the immediate predecessor task sequence, that is, each task must satisfy the priority relationship of its predecessor task, ensuring that the disassembly task is performed in the correct order;

[0036] S624, the process of the variation operation based on the direct predecessor task sequence is as follows: determine the variation individual, randomly select an individual π from the population, which represents a feasible sequence of disassembly tasks; randomly select a task, randomly select a disassembly task i in the task sequence of the individual, which will become the object of variation; select the predecessor task, find the immediate predecessor task of the disassembly task i, that is, the disassembly task j directly dependent on the disassembly task i in the disassembly sequence, otherwise it will lead to the task sequence not meeting the priority relationship; adjust the task sequence, in order to perform the variation operation, insert the disassembly task i in front of or behind the predecessor disassembly task j, thereby generating a new task sequence.

[0037] The one based on the number of worker skills multi-objective parallel waste product disassembly line optimization method, the method for updating the population by the non-dominated sorting method in S63, selecting the Pareto optimal solution set, and gradually iterating until the stop condition is met includes:

[0038] S631、In this population updating process, first of all, the individual is selected by non-dominated sorting; according to the dominance rule, the algorithm will divide the solution into multiple levels, the solution which is not dominated by other solutions is called Pareto optimal solution, and these solutions form the first layer; for the remaining solutions, according to the dominated situation, they are assigned to the next layer in turn;

[0039] Assuming there are U objective functions, if the solution x0 can dominate the solution x1, the following conditions need to be met:

[0040]

[0041] In this way, non-dominated sorting classifies the population by level, ensuring that each generation of individuals has certain superiority; At the same time, in order to ensure the diversity of solutions, the concept of crowding distance is introduced; Crowding distance selects more sparse solutions by calculating the density of individuals in the solution space, and sparse individuals are superior in crowding distance, avoiding the concentration of the population in a certain area, so as to achieve the wide distribution of solutions;

[0042] S632、On this basis, the article forms the next generation population by combining these sorted individuals with the current population; When updating, the algorithm preferentially selects those relatively sparse individuals according to the crowding distance, thereby maintaining the diversity of the population; In multi-objective optimization, the fitness of an individual is determined by multiple objective functions, and the combination of non-dominated sorting and crowding distance allows each individual to consider all objectives, thereby avoiding the limitations of single-objective optimization; This method not only retains excellent individuals, but also ensures the coverage of the solution space through diversified updates;

[0043] S633、After multiple iterations, the algorithm will gradually approach the optimal solution set, i.e. the Pareto optimal solution; Each iteration will find the current Pareto solution set through non-dominated sorting, and combine it with the crowding distance to ensure the diversity of the solution; With the iteration, the Pareto solution set is updated continuously until the predetermined stopping condition is met, such as reaching the maximum number of iterations or the solution set converges, and a balanced multi-objective optimization solution set is finally obtained; This process ensures the efficiency and accuracy of optimization.

[0044] The improved IDBA is compared with NSGA-II, NSGA-III, SPEA-II, ESPEA and MOEA / D five algorithms in actual cases, including:

[0045] S10.1、Select five products with different numbers of tasks, and randomly combine them into three different forms of product combinations;

[0046] S10.2, for all algorithms, set the population size to 100, the number of iterations to 100, the crossover probability to 0.7, the mutation probability to 0.5 times the reciprocal of the task sequence length, and the cycle time of the workstation to 50;

[0047] S10.3, using the first product combination as an experimental case, the IDBA algorithm and the NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D algorithms are compared from the perspectives of IGD metrics, HV metrics, and Epsilon-metric metrics, and the results prove that the IDBA algorithm has more advantages than the other algorithms in processing multi-objective parallel disassembly line problems;

[0048] S10.4, then, the IDBA algorithm and the NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D algorithms are compared in terms of changes in three different evaluation indicators when the iteration time and the population size change while processing actual problems;

[0049] S10.5, in order to further prove the superiority of the IDBA algorithm in processing actual problems, three box plots are used to record the values of IGD metrics, HV metrics, and Epsilon metrics;

[0050] S10.6, finally, the allocation of skilled workers and disassembly tasks in the disassembly line workstation of the first product combination experimental case is output.

[0051] The beneficial effects of the present application are:

[0052] 1. The present application can optimize the profit and skill requirement of the disassembly line: the present application proposes an improved discrete bat algorithm suitable for multi-objective parallel disassembly line balancing problems. The method optimizes the performance of the disassembly line by maximizing the disassembly profit and minimizing the required number of worker skills, thereby realizing an efficient disassembly process.

[0053] 2. The present application provides multiple solution selection options: the improved discrete bat algorithm (IDBA) in the present application is based on the Pareto dominance relationship, selects the optimal solution through non-dominated sorting, and guides the algorithm to develop towards the optimal solution. It combines two main operators: the order-preserving crossover (PPX) operator and the position-based mutation operator, ensuring that both excellent genes are preserved and local optimal solutions are jumped out in multi-objective optimization, thereby improving the global optimization performance. Through these optimization strategies, the best solution can be flexibly selected from a series of balancing schemes according to different disassembly requirements, production limitations, and resource allocation in the Pareto frontier.

[0054] 3.The IDBA algorithm of the present application can find a proper balance point among multiple goals such as disassembly rhythm, profit maximization and worker skill minimization. Experimental results show that IDBA performs well in various industrial disassembly scenarios, and the generated Pareto solution set is more evenly distributed and more convergent, and the solution set shows high flexibility and practicality in multi-objective optimization. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 A flowchart of a parallel disassembly line balancing optimization method considering the number of worker skills in an embodiment of the present application;

[0056] Figure 2 An AND-OR graph of a product to be disassembled in an embodiment of the present application;

[0057] Figure 3 A skill allocation graph in an embodiment of the present application;

[0058] Figure 4 An encoding strategy diagram in an embodiment of the present application;

[0059] Figure 5 A decoding strategy flowchart in an embodiment of the present application;

[0060] Figure 6 A diagram of the initialization process in an embodiment of the present application;

[0061] Figure 7 A diagram of the three stages of the adjustment process in an embodiment of the present application;

[0062] Figure 8 A flowchart of the improved discrete bat optimization algorithm in an embodiment of the present application;

[0063] Figure 9 A diagram comparing the Pareto frontiers of the IDBA algorithm and other five algorithms in an embodiment of the present application DETAILED DESCRIPTION

[0064] To make the technical problems, technical solutions and advantages of the present application clearer, specific embodiments will be described in detail below with reference to the accompanying drawings.

[0065] An optimization method for a parallel disassembly line balancing problem considering the number of skills possessed by workers is provided in an embodiment of the present application, which can be implemented by electronic equipment. Figure 1 A flowchart of an optimization method for a parallel disassembly line balancing problem considering the number of skills possessed by workers, the processing flow of the method of the present application includes the following steps:

[0066] S1, according to the product information and disassembly line information of the product to be disassembled and the disassembly skill information possessed by the workers, a disassembly line balancing model is constructed;

[0067] S2, construct a target function aiming at minimizing the number of skills required by workers and maximizing the net profit of product disassembly;

[0068] S3, set the constraint conditions of the disassembly line balancing optimization model;

[0069] S4, encode the disassembly tasks based on the coding mechanism of the priority relationship matrix, generate feasible solutions, and decode;

[0070] S5, solve the multi-objective disassembly line balancing optimization model by using the improved discrete bat algorithm (IDBA) and obtain the task allocation scheme of the products to be disassembled;

[0071] S6, compare the improved IDBA with NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D in actual cases.

[0072] The steps S1 to S6 will be described in detail as follows:

[0073] First, step S1 will be described in detail:

[0074] S11, the information of the products to be disassembled includes the number of disassembly tasks, the number of components, the number of parts of the products to be disassembled, the unit time disassembly cost and the recycling cost of the product components, the priority relationship between the disassembly tasks of the products, the disassembly relationship between the tasks and the components, and the disassembly time of each task on the disassembly line;

[0075] Among them, the AND / OR graph of the product information described in S11 is the AND / OR graph of the disassembly process of a real product ballpoint pen. Through the AND / OR graph of the product, the disassembly relationship between the tasks in the product can be mastered; Figure 2 S12, the disassembly line information includes the cycle period of the disassembly line, which is a parameter. In specific actual cases, it can be appropriately set according to the actual disassembly situation;

[0076] S13, all disassembly skill information of workers includes the number of disassembly skills mastered by workers in the product disassembly process;

[0077] Among them, the disassembly skill information of workers in the disassembly line described in S13 is the disassembly skill information of workers in the disassembly line. The boxes of the same color represent the same product disassembly skills, and the numbers in the boxes represent the disassembly task numbers of the products;

[0078] Figure 3 Similarly, the disassembly skill information of workers in the disassembly line described in S13 is the disassembly skill information of workers in the disassembly line. The boxes of the same color represent the same product disassembly skills, and the numbers in the boxes represent the disassembly task numbers of the products;

[0079] Similarly, the disassembly skill information of workers in the disassembly line described in S13 is the disassembly skill information of workers in the disassembly line. The boxes of the same color represent the same product disassembly skills, and the numbers in the boxes represent the disassembly task numbers of the products; Figure 3 ​The left half of the figure represents that the workers of the first workstation W1 need two disassembly skills, and the workers of W2 also need two disassembly skills, so the task distribution requires the workers of the two workstations to master a total of four skills; but the right half of the figure represents that the workers of the W1 workstation only need to master one skill, and the workers of W2 also only need to master one skill, so the task distribution only requires the workers of the two workstations to master a total of two skills. Figure 3 The right half of the figure represents that the workers of the W1 workstation only need to master one skill, and the workers of W2 also only need to master one skill, so the task distribution only requires the workers of the two workstations to master a total of two skills. The objective function for minimizing the number of skills that the workers need to master and maximizing the net profit of product disassembly is proposed, aiming to improve the efficiency of the enterprise.

[0080] Next, step S2 is described in detail:

[0081] The objective function in S2 is as follows:

[0082] maxf1, minf2;

[0083]

[0084] Next, step S4 is described in detail:

[0085] The coding mechanism based on the priority relationship matrix is used to code the disassembly tasks, generate feasible solutions, and decode;

[0086] The coding mechanism based on the priority relationship matrix in S4 includes:

[0087] S411, according to the and / or graph of the product disassembly task execution order, the priority relationship matrix of the task is obtained;

[0088] S412, according to the priority relationship matrix of the task, a vector π is introduced to represent a disassembly task sequence of the product, each element in π corresponds to a task number and is arranged in order;

[0089] S413, this vector π is a task coding method. By introducing the vector π, the system can digitally represent the disassembly sequence. For some complex disassembly tasks, the vector π ensures that each task can follow the rules of the priority relationship matrix, thereby maintaining the logical order of the tasks;

[0090] Example Figure 4 The task is real-coded;

[0091] The generation of feasible solutions for disassembly tasks in S4 includes:

[0092] S421, in the balancing problem of parallel disassembly lines, two straight-line disassembly lines are considered, and each disassembly line is assigned a single product type;

[0093] S422, generating a non-repetitive random number according to the number of disassembly tasks of the product, inserting the generated random number into the sequence to generate a feasible solution, which greatly ensures the randomness and integrity of the disassembly sequence;

[0094] The decoding of the disassembly tasks in S4 includes:

[0095] S431, distributing the generated disassembly task sequence to different workstations to ensure that the task time of the workstation does not exceed its cycle time.

[0096] In this decoding process, not only the disassembly sequence of the task is considered, but also the constraint condition of the task time is satisfied;

[0097] Example Figure 5 The complete task decoding process diagram.

[0098] Next, step S5 is described in detail:

[0099] S51, initialization and adjustment of the population;

[0100] S52, updating the individual through the cross operation based on priority reservation cross (PPX) and the mutation operation based on the direct pre-task sequence;

[0101] S53, updating the population by non-dominated sorting, selecting the Pareto optimal solution set, and gradually iterating until the stop condition is met, including:

[0102] S54, obtaining a disassembly line balancing strategy that satisfies the balance constraint condition;

[0103] The initialization and adjustment of the population in S51 includes:

[0104] S511, each individual in the population represents a feasible disassembly sequence, which is generated by the random insertion method as described in the initialization process and the encoding process, ensuring the randomness and diversity of each individual;

[0105] S512, the specific method is to randomly insert the disassembly task sequence of product II into the disassembly task sequence of product I;

[0106] Example Figure 6 The population initialization process;

[0107] S513, when generating a feasible disassembly sequence of a parallel disassembly line, it is necessary to ensure the disassembly priority relationship between the disassembled products on the parallel disassembly line;

[0108] Taking a simple product parallel disassembly line of two products as an example, the disassembly sequence generated for each product by the random integer method may not be feasible. Therefore, it is necessary to adjust the generated sequence to make it a feasible disassembly sequence;

[0109] S514, the adjustment process is divided into three stages: the addition operation, the missing tasks are added to the existing infeasible sequence according to the priority relationship; the removal operation, the conflicting tasks in the task sequence are removed (the conflicting tasks refer to the tasks that cannot exist in the same disassembly sequence at the same time); the exchange operation, after the completion of the previous two stages, if some tasks still do not meet the priority relationship, the order of the tasks is adjusted;

[0110] Example Figure 7 to initialize the three stages of the adjustment process;

[0111] Among them, the updating of the individual according to the priority-based reserved crossover operator (PPX) and the mutation operation based on the direct predecessor task sequence in S2 includes:

[0112] S521, the crossover operation based on the priority-based reserved crossover operator, the purpose is to inherit the excellent genes in the parent when updating the individual, and to ensure that the priority relationship of the disassembly tasks is not damaged;

[0113] S522, the process of the priority-based reserved crossover operation is as follows: first, input two individuals, respectively, the current best individual π f and the individual π m randomly selected from the current Pareto solution set, and a randomly generated mask sequence MASK; if the mask bit is 1, the corresponding parent individual π m element is filled into the new individual π new ; if the mask bit is 0, the corresponding parent individual π f element is filled into the new individual π new ; finally output: the generated new individual π new ;

[0114] S523, the mutation operation based on the direct predecessor task sequence, which can make local adjustments to the individual while maintaining the basic structure of the individual, thereby improving the diversity of the solution and helping the algorithm jump out of the local optimum. In the optimization method, the mutation operation is based on the immediate predecessor task sequence, that is, each task must satisfy the priority relationship of its predecessor task, ensuring that the disassembly tasks are performed in the correct order;

[0115] S524, the process of the variation operation based on the direct pre-sequence task sequence is as follows: determining a variation individual, randomly selecting an individual pi from the population, the individual representing a feasible sequence of disassembly tasks; randomly selecting a task, randomly selecting a task i in the task sequence of the individual, the task i being the object of the variation; selecting a pre-sequence task, finding an immediately pre-sequence task j of the task i, that is, a task directly dependent on the task i in the disassembly sequence (the task j must be completed before the task i), otherwise the task sequence will not meet the priority relationship; adjusting the task sequence, in order to perform the variation operation, inserting the task i in front of or behind the pre-sequence task j, thereby generating a new task sequence;

[0116] Wherein, the population is updated by the non-dominated sorting method, the Pareto optimal solution set is selected, and the iteration is gradually performed until the stop condition is met in S3, comprising:

[0117] S531, in the population updating process, the individual is selected by non-dominated sorting. According to the dominance rule, the algorithm divides the solution into multiple levels, and the solution which is not dominated by other solutions is called Pareto optimal solution, and these solutions form the first layer. For the remaining solutions, according to the dominated condition, they are assigned to the next layer in turn;

[0118] Suppose there are U objective functions, if the solution x0 can dominate the solution x1, the following conditions need to be met:

[0119]

[0120] In this way, the non-dominated sorting classifies the population by levels, ensuring that each generation of individuals has certain superiority;

[0121] At the same time, in order to ensure the diversity of the solution, the concept of crowding distance is also introduced. The crowding distance selects more sparse solutions by calculating the density of individuals in the solution space, and the sparse individuals are superior in the crowding distance, avoiding the concentration of the population in a certain area, so as to realize the wide distribution of the solution;

[0122] S532, on this basis, the article forms the next generation population by combining these sorted individuals with the current population. When updating, the algorithm preferentially selects those relatively sparse individuals according to the crowding distance, so as to maintain the diversity of the population. In multi-objective optimization, the fitness of an individual is determined by multiple objective functions, and the combination of non-dominated sorting and crowding distance enables each individual to consider all objectives, thereby avoiding the limitations of single-objective optimization. This method not only retains excellent individuals, but also ensures the coverage range of the solution space through diversified updating;

[0123] S533、After multiple iterations, the algorithm gradually approaches the optimal solution set, i.e. the Pareto optimal solution. In each iteration, the current Pareto solution set is found by non-dominated sorting, and combined with the crowding distance to ensure the diversity of solutions. As the iteration proceeds, the Pareto solution set is updated until the predetermined stopping condition is met, such as reaching the maximum number of iterations or the solution set converges, and finally a balanced multi-objective optimization solution set is obtained. Figure 8 Flowchart of the IDBA algorithm.

[0124] Next, step S6 is described in detail:

[0125] The improved IDBA described in S6 is compared with NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D in actual cases, including:

[0126] S61、Select five products with different task numbers, and randomly combine them into three different forms of product combinations;

[0127] S62、For all algorithms, set the population size to 100, the number of iterations to 100, the crossover probability to 0.7, the mutation probability to 0.5 times the inverse of the task sequence length, and the workstation cycle time to 50;

[0128] S63、Using one of the product combinations as an experimental case, compare IDBA algorithm with NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D from the perspectives of IGD metric, HV metric, and Epsilon-metric metric;

[0129] S64、In multi-objective optimization problems, the Pareto frontier can intuitively show the distribution of the target values obtained by the algorithm, for example Figure 9 The Pareto frontier data graph obtained by each algorithm. The X-axis (f1) represents the target function disassembly profit, and the Y-axis (f2) represents the target function skill number;

[0130] S65、For example Figure 9 In the figure, it can be observed that high-quality solution sets are distributed in the lower right part of the graph, thus showing that the convergence degree and uniformity of the Pareto solution set obtained by IDBA are better;

[0131] S66、Finally, output the allocation of skill workers and disassembly tasks in the disassembly line workstation for the product combination case applied in the experiment.

[0132] In the exemplary embodiment, a computer readable storage medium, such as a memory including instructions executable by a processor in a terminal, is also provided to implement the optimization method for the multi-objective parallel disassembly line balancing problem. The optimization method described in the present application, including the steps of task allocation, skill minimization and profit maximization, can be implemented by a computer program.

[0133] For example, the computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk or an optical data storage device, etc. Those of ordinary skill in the art can understand that all or part of the steps of the above-described embodiments can be completed by hardware or instructed to relevant hardware by software programs, and these programs can be stored in a computer readable storage medium. The above storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0134] The above description is only the preferred embodiment of the present application, and is not intended to limit the scope of the present application. Any modification, equivalent replacement or improvement within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for optimizing a waste product dismantling line based on multiple objectives and parallel processing of worker skill levels, characterized in that, The method The process includes the following: S1 constructs a disassembly line balance model based on the product information of the product to be disassembled, the disassembly line information, and the disassembly skills of the workers; S2 constructs an objective function with the goal of minimizing the number of skills required by workers and maximizing the net profit from product dismantling; S3 sets the constraints for the disassembly line balancing optimization model; S4 uses a priority relation matrix-based encoding mechanism to encode, generate feasible solutions, and decode the disassembly task; S5 uses the improved Discrete Bat Algorithm (IDBA) to solve the multi-objective disassembly line balance optimization model and obtains the task allocation scheme for the products to be disassembled. S6 compares the improved IDBA with five algorithms—NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D—in practical cases. The information of the product to be disassembled includes the number of disassembly tasks, the number of components, the number of parts, the unit time disassembly cost and recycling cost of the product components, the priority relationship between the product disassembly tasks, the disassembly relationship between tasks and components, and the disassembly time of each task on the disassembly line. The disassembly line information includes the disassembly line cycle time; the worker's disassembly skills information includes the number of disassembly skills the worker has mastered during the product disassembly process. The encoding mechanism based on the priority relation matrix encodes, generates feasible solutions, and decodes the disassembly task, including: S51. Obtain the priority relationship matrix of tasks based on the AND-OR graph of task execution order in the product; S52. Based on the task priority matrix, introduce a vector... To represent a sequence of disassembly tasks for a product. Each element in the vector corresponds to a task number and is arranged sequentially; It is a task coding method; by introducing vectors The system can digitally represent the disassembly sequence; for some complex disassembly tasks, vector... This ensures that each task follows the rules of the priority matrix, thus maintaining the logical order of the tasks; S53. Generate feasible solutions using the sequential insertion method; insert the disassembly task sequence of one product into the task sequence of another product to ensure that the disassembly tasks are assigned sequentially, meeting the priority and time requirements of the tasks; S54. Distribute the generated task disassembly sequence to different workstations, ensuring that the task time of each workstation does not exceed its cycle time. In this decoding process, not only the task disassembly order must be considered, but also the task time constraints must be met.

2. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 1, is characterized in that... The method aims to minimize the number of skills required from workers and maximize the net profit from product dismantling.

3. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 1, is characterized in that... The improved discrete bat optimization algorithm is used to solve the multi-objective disassembly line balance optimization model, and the resulting task allocation scheme for the products to be disassembled includes: S61. Population initialization and adjustment; S62. Individuals are updated through a priority-based crossover operation (PPX) and a mutation operation based on the direct preceding task sequence. S63. Update the population using a non-dominated sorting method, select the Pareto optimal solution set, and iterate step by step until the stopping condition is met; S64. Obtain the disassembly line balancing strategy that satisfies the balancing constraints.

4. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 3, is characterized in that... The initialization and adjustment of the population in S61 includes: S611. Each individual in the population represents a feasible disassembly sequence; the population is generated by random insertion to ensure the randomness and diversity of each individual; specifically, the disassembly task sequence of product II is randomly inserted into the disassembly task sequence of product I. S612. When generating a feasible disassembly sequence for a parallel disassembly line, it is necessary to ensure that the disassembly tasks of each product follow their priority relationship; however, randomly generated sequences may not necessarily meet these constraints, so further adjustments are necessary. S613. The adjustment process is divided into three stages: addition operation, which adds the missing tasks to the existing infeasible sequence according to their priority; removal operation, which removes conflicting tasks from the task sequence; and swap operation, which adjusts the task order if some tasks still do not meet the priority relationship after the first two stages are completed.

5. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 3, is characterized in that... The method for updating individuals in S62 through priority-based crossover operation PPX and mutation operation based on direct preceding task sequence includes: S621. Priority-preserving crossover operation aims to ensure that individuals can inherit superior genes from their parents when updating them, and to ensure that the priority relationship of dismantling tasks is not disrupted. S622. The process of crossover operation based on priority preservation is as follows: First, input two individuals, which are the current best individuals. and individuals randomly selected from the current Pareto solution set And a randomly generated mask sequence MASK; if the mask bit is 1, then the corresponding parent individual The elements are filled into the new individual. In the middle; if the mask bit is 0, then the corresponding parent individual will be... The elements are filled into the new individual. Middle; Final output: The newly generated individual ; S623. The mutation operation based on the immediate preceding task sequence can make local adjustments to individuals while maintaining their basic structure, thereby improving the diversity of solutions and helping the algorithm escape local optima. In this optimization method, the mutation operation is based on the immediate preceding task sequence, that is, each task must satisfy the priority relationship of its preceding tasks to ensure that the disassembly tasks are performed in the correct order. S624. The mutation operation based on the direct preceding task sequence is as follows: Determine the mutated individual, and randomly select an individual from the population. This individual represents a feasible sequence of disassembly tasks; a task is randomly selected from the individual's task sequence. This task will become the object of mutation; select the preceding task and find the disassembly task. The immediate preceding task, i.e., the disassembly task. The disassembly task directly depends on the disassembly sequence. Otherwise, the task order will not conform to the priority relationship; adjust the task order, and in order to perform the mutation operation, disassemble the tasks. Reinsert it before or after the preceding task to create a new task order.

6. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 3, is characterized in that... In step S63, updating the population using a non-dominated sorting method, selecting the Pareto optimal solution set, and iterating step by step until the stopping condition is met includes: S631. In the population update process, individuals are first selected by non-dominated sorting. According to the domination rule, the algorithm divides the solutions into multiple levels. The solutions that are not dominated by other solutions are called Pareto optimal solutions, and these solutions form the first layer. For the remaining solutions, they are assigned to the next layer in turn according to their domination status. Suppose there are U objective functions, if the solutions... Can control solutions Then the following conditions must be met: ； Non-dominated sorting classifies the population hierarchically, ensuring that individuals in each generation have a certain degree of superiority. At the same time, in order to ensure the diversity of solutions, the concept of crowding distance is introduced. Crowding distance selects sparser solutions by calculating the density of individuals in the solution space. Sparse individuals are dominant in crowding distance, avoiding the population from being concentrated in a certain area, thereby achieving a wide distribution of solutions. S632. Based on this, the next generation of population is formed by combining these sorted individuals with the current population. During the update, the algorithm prioritizes sparser individuals based on crowding distance, thereby maintaining the diversity of the population. In multi-objective optimization, the fitness of an individual is determined by multiple objective functions. The combination of non-dominated sorting and crowding distance allows each individual to comprehensively consider various objectives, thereby avoiding the limitations of single-objective optimization. This method retains excellent individuals and ensures the coverage of the knowledge space through diversified updates. S633. After multiple iterations, the algorithm gradually approaches the optimal solution set, i.e., the Pareto optimal solution. In each iteration, the current Pareto solution set is found through non-dominated sorting, and crowding distance is combined to ensure the diversity of solutions. As the iteration proceeds, the Pareto solution set is continuously updated until the predetermined stopping conditions are met, such as reaching the maximum number of iterations or when the solution set converges, and finally a balanced multi-objective optimization solution set is obtained. This process ensures the efficiency and accuracy of optimization.

7. The method for optimizing a waste product dismantling line based on multiple objectives and parallel operation of worker skill levels, as described in claim 1, is characterized in that... The comparison of the improved IDBA with five algorithms—NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D—in practical cases includes: S10.1 Select five products with different task numbers and randomly combine them in pairs into three different product combinations; S10.2 For all algorithms, set the population size to 100, the number of iterations to 100, the crossover probability to 0.7, the mutation probability to 0.5 times the reciprocal of the task sequence length, and the workstation cycle time to 50. S10.3 Using the first product combination as an experimental case, the IDBA algorithm was compared with five other algorithms, namely NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D, from the perspectives of IGD metric, HV metric, and Epsilon-metric. The results showed that the IDBA algorithm has more advantages than the other algorithms in dealing with multi-objective parallel disassembly line problems. S10.4 Next, the changes of three different evaluation indicators of the IDBA algorithm and five other algorithms (NSGA-II, NSGA-III, SPEA-II, ESPEA, and MOEA / D) in handling practical problems were compared with the changes in iteration time and population size. S10.5 To further demonstrate the superiority of IDBA in addressing real-world problems, three box plots were used to record the values ​​of the IGD metric, HV metric, and Epsilon metric. S10.

6. Final output: The allocation of skilled workers and dismantling tasks in the dismantling line workstation of the first product combination experimental case.

Images