Method for mixed linear local disassembly optimization for multi-retired product recycling

Abstract

The application relates to a mixed linear partial destruction disassembly optimization method for multi-retired product recycling, relates to the technical field of disassembly line layout, and mainly comprises the following steps: collecting product disassembly information, determining a target function, initializing parameters and product information, generating an initial firefly population, performing Pareto screening on the initial firefly population to obtain dominant individuals, performing firefly algorithm iteration calculation, judging whether the algorithm is restarted, selecting different ways to screen and calculate non-inferior solutions according to the restart condition of the algorithm, updating an external file according to the non-inferior solution, judging whether the maximum iteration number is reached, returning to continue iteration if the maximum iteration number is not reached, and otherwise outputting a result; the improved multi-objective firefly algorithm provided by the application establishes a mixed integer programming model for mixed lines of multi-retired product recycling of different types, mainly optimizes the number of combined workstations, smoothness, safety and disassembly profit, and provides a feasible theoretical guidance for disassembly line layout optimization.

Description

Technical Field

[0001] This invention relates to the field of facility layout technology, specifically to a hybrid linear partial destruction dismantling optimization method for the recycling of multiple decommissioned products. Background Technology

[0002] With rapid industrialization and the surge in electronic products, product lifecycles have shortened significantly, leading to faster product upgrades and increased waste generation. This waste poses a dual threat to the environment and resources. To achieve sustainable and environmentally friendly waste management, effective recycling and remanufacturing measures are essential. By recovering valuable resources from waste, we can reduce pollution from hazardous substances while also generating substantial benefits. Remanufacturing and dismantling, as the first step in waste remanufacturing and recycling, plays a crucial role in the success of resource recovery. However, traditional dismantling methods face numerous challenges, especially in large-scale, multi-product dismantling scenarios. Traditional dismantling models and workstation layouts often fail to meet the growing recycling demands of the green circular economy and the increasingly complex requirements of dismantling tasks.

[0003] Existing dismantling methods typically process similar end-of-life products on the same production line. However, the type, quality, and quantity of end-of-life products often vary significantly during dismantling. Unfortunately, current workstation layouts are usually designed for single product types, mostly employing a linear layout. In contrast, a dual-sided layout offers greater flexibility, reduces the movement of large, idle products, and thus improves dismantling efficiency. However, this method alone cannot meet the complex and diverse task requirements of large-scale recycling scenarios, necessitating the use of multiple dismantling lines. Therefore, this research aims to improve existing layouts by designing a flexible and efficient parallel dual-sided layout to meet the diverse dismantling needs of retired products, thereby enhancing the system's flexibility and adaptability. Summary of the Invention

[0004] Therefore, the main objective of this invention is to provide a hybrid local destruction dismantling line balancing optimization method for recycling multiple retired electronic products and automobile engines for green remanufacturing based on the Improved Multi-Objective Firefly Algorithm (IMOFA). This method aims to obtain a comprehensive and better solution for simultaneously dismantling multiple types of scrapped home appliances and engines, effectively solving the problem of non-dismantlable components in multi-product recycling scenarios, improving dismantling line efficiency, supporting green remanufacturing, and ultimately improving enterprise dismantling profits and the environment.

[0005] The technical solution of this invention is a hybrid linear localized destruction dismantling optimization method for recycling multiple decommissioned products, comprising the following steps:

[0006] Step S1: Collect product disassembly information and determine the objective function with disassembly efficiency, production line length, safety risks, and disassembly profit as optimization objectives;

[0007] Step S2: Initialize parameters and product information, generate an initial firefly population, perform Pareto screening on the initial firefly population to identify dominant individuals, and perform iterative calculations using the firefly algorithm.

[0008] Step S3: Determine whether to restart the algorithm, and select different methods to filter and calculate non-dominated solutions based on the restart status of the algorithm;

[0009] Step S4: Update the external file based on the non-dominated solution, and determine whether the maximum number of iterations has been reached. If not, return to step S3 to continue iterative calculation. If the maximum number of iterations has been reached, output the result in the external file.

[0010] The technical effects of this invention are:

[0011] This invention proposes an improved multi-objective firefly algorithm (IMOFA) to address the parallel bilateral locally destructive dismantling line balancing problem for green remanufacturing. By establishing a mixed-integer programming model for a hybrid line type accommodating the recycling of various types of retired products, the model focuses on optimizing the number of workstations, smoothness, security, and dismantling profits to promote resource conservation and environmental protection, providing feasible theoretical guidance for the optimal layout of dismantling lines. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below.

[0013] Figure 1 This is a schematic diagram of the hybrid dismantling line layout optimized based on the improved multi-objective firefly algorithm proposed in this invention.

[0014] Figure 2 This is a flowchart of the improved multi-target firefly algorithm in this invention;

[0015] Figure 3 This is a schematic diagram of the three-point intersection operation in this invention;

[0016] Figure 4 This is a schematic diagram of the mutation operation in this invention;

[0017] Figure 5 This is a priority relationship diagram of different engines in the case of dismantling retired engines in this invention;

[0018] Figure 6 Box plots of the five multi-objective optimization algorithms in this invention under three multi-objective indices;

[0019] Figure 7This is a Gantt chart of dismantling scheme No. 1 among the four single-objective minimum dismantling schemes obtained in the dismantling process of a decommissioned engine based on the improved multi-objective firefly optimization algorithm in this invention;

[0020] Figure 8 This is a Gantt chart of dismantling scheme No. 2 among the four single-objective minimum dismantling schemes obtained in the dismantling process of a decommissioned engine based on the improved multi-objective firefly optimization algorithm in this invention;

[0021] Figure 9 This is a Gantt chart of dismantling scheme No. 3, which is one of the four single-objective minimum dismantling schemes obtained in the dismantling process of a decommissioned engine based on the improved multi-objective firefly optimization algorithm in this invention.

[0022] Figure 10 This is a Gantt chart of dismantling scheme No. 5, which is one of the four single-objective minimum dismantling schemes obtained in the dismantling process of a decommissioned engine based on the improved multi-objective firefly optimization algorithm in this invention. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings.

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example:

[0026] A hybrid linear localized destruction dismantling optimization method for recycling multiple decommissioned products includes the following steps:

[0027] Step S1: Collect product disassembly information and determine the objective function with disassembly efficiency, production line length, safety risks, and disassembly profit as optimization objectives.

[0028] Traditional production lines only consider the basic attributes of the product during disassembly, such as hazardous attributes, demand attributes, and non-existent attributes, neglecting the issue of component non-disassembly caused by corrosion, aging, and deformation. Furthermore, due to the large variety and quantity of retired products, this invention proposes a hybrid disassembly line layout to simultaneously disassemble different models of retired products, such as... Figure 1As shown, two double-sided production lines are run in parallel. The double-sided production lines can be disassembled on both the left and right sides of the line. Running two double-sided production lines in parallel significantly improves the efficiency of disassembling multiple products. Simultaneously, the right side of the first line and the left side of the second line form a common workstation, reducing the number of operators and thus lowering disassembly costs. Therefore, the objective function to be satisfied in this invention is:

[0029]

[0030] Equation (1) is the weighted sum and minimization of the number of workstations in a workstation combination and the number of workstations.

[0031]

[0032] Equation (2) indicates that the smoothing index can be reduced by minimizing the smoothing index by reducing line length and ensuring a balance of idle time.

[0033]

[0034] Equation (3) indicates that dangerous tasks should be handled as early as possible. Since a destructive dismantling mode is introduced and dangerous tasks exist during the dismantling process, there may be safety hazards during the dismantling process. Therefore, it is necessary to dismantle dangerous tasks as early as possible.

[0035]

[0036] Equation (4) represents the maximization of dismantling profits, including the dismantling revenue and dismantling costs of tasks under different dismantling modes, the fixed start-up cost of each workstation, and the cost of additional handling of dangerous tasks.

[0037] Based on this, the objective function needs to satisfy the following constraints:

[0038]

[0039] The constraints in equations (8) and (9) indicate that all tasks need to be disassembled:

[0040]

[0041] The constraints in equations (10) and (11) indicate that each task needs to be assigned to its respective production line:

[0042]

[0043] The constraints in equation (12) indicate that some tasks are directionally restricted. Therefore, tasks with left-side directional restrictions can only be assigned to the left, while tasks with right-side directional restrictions can only be assigned to the right.

[0044]

[0045] Equations (13) and (14) indicate that not all tasks can be disassembled in a destructive manner; dangerous tasks and demanding tasks, due to their inherent properties, can only be disassembled non-destructively.

[0046]

[0047] Equation (15) indicates that, except for hazardous and demand-based tasks, other tasks can arbitrarily choose their dismantling mode. When the feasibility value of the destructive dismantling of the task is e... i Greater than a defined threshold e max When r indicates that the task requires destructive disassembly, then r i =1, otherwise a disassembly mode will be randomly selected:

[0048]

[0049] Equations (16) and (17) indicate that tasks assigned to the same workstation must wait until their predecessor tasks are completed before they can be dismantled.

[0050]

[0051] Equation (18) indicates that each process must be completed within the cycle time of the specified workstation:

[0052]

[0053] Equations (19) and (20) indicate that in order to establish and Given the relationship that task i in production line h is assigned to the p side of the kth workstation before k, this constraint restricts task i and task s to be assigned to the same side of the same workstation:

[0054]

[0055] Equation (21) indicates that the disassembly of subsequent tasks can only begin after the preceding tasks assigned to the same-side workstations have been completed:

[0056]

[0057] Equation (22) indicates that the disassembly task can only begin after its preceding task has been completed:

[0058]

[0059] Equation (23) represents the time limit for tasks on public workstations:

[0060] A·(1-y jik )+w hi ≥w (3-h)j +t (3-h)j,i∈I h ,j∈I (3-h) ,k∈M (23)

[0061] Equations (24) and (25) represent and y ijk The relationship between them:

[0062]

[0063] Equations (26) to (28) represent the upper and lower limits on the number of workstations:

[0064]

[0065] Equation (29) indicates that the workstations are turned on sequentially:

[0066]

[0067] Equations (30) to (33) represent the restrictions on opening the combined workstation:

[0068]

[0069] In the above formulas, m is the maximum number of available workstations; M is the set of workstation numbers, M = {1, 2, ..., m}; γ is the production line weighting coefficient, γ1 = 100, γ2 = 1; S k Enable the variable S for the combined workstation k =1 indicates that the k-th combined workstation is on, and S is off when the workstation is off. k =0; k is the workstation number index; p represents the position variable of the task assigned to the workstation, P is the side set, P = {p|1,2}, where p = 1 represents the left side, p = 2 represents the right side, p = h indicates that the side p of the current workstation corresponds to the production line h, p = 3-h indicates that the side of the current workstation is opposite to the production line; 3-h represents another production line corresponding to the current production line h, h = {1,2}, when h = 1, 3-h = 2, when h = 2, 3-h = 1. This indicates whether the p-side of the common workstation k between the two production lines is open; if open, then... otherwise h is the double-sided production line number, H is the set of double-sided production line numbers, H = {1, 2}; CT is the cycle time. This indicates whether the p-side of workstation k belonging to this production line in the common workstation is enabled for the h-th production line. If enabled, then... otherwise n h Let be the total number of disassembly tasks on the h-th production line; i,j,s are the disassembly task numbers. Assign variables to the task. This indicates that the i-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... t hi L represents the disassembly time of task i on production line h; k Indicates whether the public workstation is enabled, L k =1 indicates that it is turned on, L k =0 indicates that it is not enabled; h i This indicates that task i has a dangerous attribute, h i =1 indicates that task i is dangerous, h i =0 indicates that task i is not dangerous; G hi C represents the dismantling revenue that task i obtains on its production line; hi Let A be the cost of disassembling task i on its production line; A is a maximum real number. Enable variables for workstations on the h-th production line. If the k-th workstation on the h-th production line is enabled, then... otherwise C h This represents the additional dismantling cost required to handle hazardous task i; r i Let r be the disassembly mode variable for task i. If task i is a destructive disassembly, then r i =1; otherwise r i =0; The disassembly time required for Task i to utilize destructive disassembly. The disassembly time required for task i to utilize non-destructive disassembly; h Let I be the set of dismantling task numbers for the h-th production line. h ={1,2,...,n h}; To utilize the dismantling benefits generated from destructive dismantling for task i; To utilize the dismantling benefits generated by non-destructive dismantling for task i; The dismantling costs incurred by utilizing destructive dismantling for task i; The disassembly cost incurred by utilizing non-destructive disassembly for task i; Assign variables to the task. The time indicates that the i-th task of the 3-h production line will be assigned to the p-side of the k-th workstation; otherwise... Assign variables to the task. When the time is specified, the i-th task of the h-th production line is assigned to the left side of the k-th workstation; otherwise... Assign variables to the task. When the i-th task of the 3-h production line is assigned to the left side of the k-th workstation, otherwise... T h1 T represents the set of disassembly task numbers on the left side of the workstation on the h-th production line. h2 This is the set of dismantling task numbers on the right side of the workstation on the h-th production line; Assign variables to the task. When the task is assigned, the i-th task of the h-th production line is assigned to the right side of the k-th workstation; otherwise... DS represents the set of tasks with demand; HS represents the set of tasks with hazardous attributes; e i e represents the feasibility value for destructive dismantling of task i; max To maximize the feasibility of dismantling, e max =0.5; To decompose the sequence variables for the task, This indicates that task i and task j are simultaneously assigned to the p-th side of combined workstation k on the h-th production line, with task i assigned before task j; otherwise, t hs The disassembly time of task s on the h-th production line; I (3-h) This is the set of dismantling task numbers for the other production line corresponding to the current production line h; t (3-h)s This indicates the disassembly time of task s on another production line corresponding to the current production line h; y sik To remove the sequence variable y from the public workstation sik =1 indicates that tasks i and j are assigned to the right side of production line h or the left side of production line 2 at workstation k, and not on the same side. Task i is assigned before task j. Otherwise, y sik =0; Assign variables to the task. This indicates that the j-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... This represents the disassembly order variable. Assign variables to the task. This indicates that task i and task j are simultaneously assigned to the p-th side of the combined workstation k, and task i is assigned before task j; otherwise... Assign variables to the task. This indicates that task i and task j are simultaneously assigned to the p-th side of the combined workstation k, and task j is assigned before task i; otherwise... Assign variables to the task. This indicates that the j-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... w hiw represents the start time of task i on the h-th production line; hj t represents the start time of task j on the h-th production line; hj Let the disassembly time of task j on production line h be _j_. Let i represent the priority relationship between task i and task j in the h-th production line; Assign variables to the task, where 3-p represents the side opposite to the current side p. This indicates that the j-th task of the h-th production line is assigned to the opposite face of the p-th side of the k-th workstation; otherwise... Assign variables to the task. This indicates that the j-th task in the other production line number corresponding to the current production line h will be assigned to the p side of the k-th workstation; otherwise... y ijk y is the dismantling order variable for public workstations. ijk =1 indicates that tasks i and j are assigned to the right side of production line 1 or the left side of production line 2 of workstation k, respectively, and are not on the same side. Task i is assigned before task j. Otherwise, y ijk =0; y jik y is the dismantling order variable for public workstations. jik =1 indicates that tasks i and j are assigned to the right side of production line 1 or the left side of production line 2 of workstation k, respectively, and are not on the same side. Task j is assigned before task i. Otherwise, y jik =0; Let h be the switch variable for the production line. If one side of the h-th production line is turned on, then... otherwise This indicates whether the variable is enabled on the left side of the first production line; To indicate whether the variable on the right side of the second production line is enabled; c m This indicates the cost of starting up the workstation;

[0070] Step S2: Initialize parameters and product information, generate an initial firefly population, perform Pareto screening on the initial firefly population to identify dominant individuals, and perform iterative calculations using the firefly algorithm.

[0071] The initial firefly population is constructed using a real-number encoding method, where each feasible task disassembly sequence is represented as a firefly individual. Additionally, an extra encoding is used to represent the task disassembly pattern. Relevant parameters include the target value of the firefly population, the number of non-dominated solutions to be output (N0), and the mutation probability (P). d wait.

[0072] Step S3: Determine whether to restart the algorithm, and select different methods to filter and calculate non-dominated solutions based on the restart status of the algorithm;

[0073] In the traditional firefly algorithm, the position update process may converge too quickly, easily leading to the algorithm getting trapped in local optima. Therefore, if... Figure 2 As shown, this invention introduces a restart strategy based on the firefly algorithm. When the optimization process of the algorithm stagnates for a long time, a restart strategy is applied to the population. By recording the non-dominated solution set of each generation, if the solution set remains unchanged for 5 generations, a new firefly population is randomly generated and its target value is calculated. The current firefly population is replaced with a population of fireflies with higher brightness, and then a mutation operation is performed. To ensure that the optimal solution is not lost, after using the new generation of population generated by the new fireflies, the algorithm mixes the new population with the non-dominated solution set of the previous generation and performs Pareto selection. Since the problem is a multi-objective optimization problem, this invention uses the hypervolume index (HV) value to evaluate the solution set of the population. When the HV value remains unchanged, it indicates that the solution set is unchanged.

[0074] If the algorithm does not restart, it selects each non-Pareto dominant individual in the initial population and performs a three-point crossover operation with a randomly selected current Pareto dominant individual. Then, it performs a mutation operation on the crossover individuals in the firefly population. Finally, it performs a screening calculation on the firefly population to obtain a non-dominated solution.

[0075] The three-point cross-operation method adopted when not restarting is mainly as follows: Figure 3 As shown, after selecting the current dominant individual, each non-Pareto dominant individual in the population undergoes a three-point crossover operation with a randomly selected current Pareto dominant individual, causing the population individuals to move towards the location of the current dominant individual in the following directions. In the parent firefly individual sequence X... 1 X 2 Three points m1, m2, and m3 are randomly generated. Using these three points as boundaries, four gene fragments can be obtained. Then, two of these fragments are randomly selected as location information fragments. Excluding the location information fragments, the parent X... 1 The sequence order remains unchanged, but individual X is removed. 1 The task of decomposing the location information fragment in the middle and finding the parent X 2 The task order corresponding to the location information fragments is copied and filled into X. 1 The position of the vacancy is determined to obtain the new position New_X. 1 Similarly, New_X can be generated. 2 .

[0076] The mutation operation will be performed regardless of whether a restart occurs. Its main operations are as follows: Figure 4 As shown: Assume a firefly's location is individual X. 3Generate a random number d. If d <P d Then, a mutation operation is performed on the firefly's position. Task 15 is randomly selected as the search starting point, and then the tasks that follow are determined according to the task priority relationship, so that the movable range of task 8 is [5,9,2,11,12,3,6]. Finally, a local movement search is performed on task 15 to obtain 7 new solutions, thus realizing the mutation of the original individual.

[0077] Step S4: Update the external file based on the non-dominated solution, and determine whether the maximum number of iterations has been reached. If not, return to step S3 to continue iterative calculation. If the maximum number of iterations has been reached, output the result in the external file.

[0078] When the non-dominated solution N q If the number of solutions exceeds the number of non-dominated solutions N0 to be output, update the external file; otherwise, select the solution with N0 non-dominated solutions by ranking by crowding distance and use it to update the external file.

[0079] The following is a demonstration of the actual operation of this embodiment:

[0080] Program testing environment:

[0081] The simulation computing environment consisted of an Intel(R) Core(TM) i5-12400 CPU @ 2.5GHz, 16GB RAM, and a Windows 10 Pro operating system, and was programmed and run using MATLAB R2016a.

[0082] 1. Case studies of discarded mobile phones and laptops:

[0083] To verify the correctness and effectiveness of the proposed model and IMOFA, two different products were introduced: a laptop computer with 8 tasks and a scrapped mobile phone with 10 tasks. Specific disassembly information is shown in Table 1 below.

[0084] Table 1. Information on the dismantling of laptops and scrapped mobile phones

[0085]

[0086] Based on the optimization objectives proposed in this invention, mathematical programming was used to solve for the precise values ​​corresponding to each objective. Then, MATLAB R2022a software was used to solve the above cases under the same configuration environment using GUROBI and IMOFA respectively. The algorithm was run independently 10 times, and the result of one solution was taken separately, as shown in Table 3.

[0087] Table 2 shows the calculation results of GUROBI and IMOFA for laptop and mobile phone teardown.

[0088]

[0089] As shown in Table 2, for the recycling cases of different models of retired laptops and scrapped mobile phones, the computation time of IMOFA is slightly shorter than that of GUROBI for the fourth optimization objective. The difference in computation time for different objective functions in GUROBI is due to the different difficulty of solving each objective. However, IMOFA can find multiple non-dominated solutions at once, and the solutions include all the optimal values ​​obtained by the exact solution, thus verifying the correctness of the model and algorithm proposed in this paper.

[0090] 2. Case studies of dismantling two different models of retired engines:

[0091] To further verify the superiority of the proposed algorithm in solving this problem, this invention takes a decommissioned engine dismantling line of a dismantling company in Southwest China as an example, and uses the proposed model and method to optimize the performance of the dismantling line to verify the proposed method.

[0092] The two different engine models have 34 and 37 components respectively, and their priority relationship diagrams are as follows: Figure 5 As shown in Table 3, the specific disassembly data is listed.

[0093] Table 3 Data information for each dismantling task of retired engines

[0094]

[0095]

[0096] Test results:

[0097] Based on the actual disassembly of the engine, a predetermined cycle time CT = 740s was set. After running IMOFA ten times for each parameter combination, the average HV value was calculated. Experimental results showed that the optimal parameters were MI = 150, WN = 500, and Pd = 0.85. After determining the optimal parameters, IMOFA was run ten times with the optimal parameter combination. To further verify the performance of the proposed algorithm in solving the proposed model, WCA, WOA, NSGA-II, and PSO were selected for comparison. To ensure fairness in the algorithm comparison, each algorithm used the same number of iterations and population size. The running deadline for all algorithms was 1000s. These five algorithms were run independently 10 times within a unified time period. The maximum HV value and Pareto solution obtained by each algorithm were recorded, and three multi-objective indicators—HV, IGD, and SM—were introduced to evaluate the algorithm's performance.

[0098] Figure 6This is a multi-objective performance chart showing five algorithms for solving a hybrid linear localized damage dismantling line balancing problem for the recycling of multiple decommissioned products. From the HV (High Value) metric, IMOFA significantly outperforms the other algorithms, especially in achieving minimum overlap. IMOFA's median is much lower than other algorithms, and it doesn't exhibit extreme values, indicating that it effectively reduces the overlap of dismantling line tasks, thereby optimizing overall dismantling efficiency. From the IGD (Increase in Displacement) metric, IMOFA also performs well, with a much lower IGD value than other algorithms. This means that IMOFA converges better in the objective space, generating a more uniform distribution of solutions that are closer to the true optimal solution when solving multi-objective optimization problems, without significant deviations. On the SM (Simplified Methods) metric, IMOFA is slightly inferior to NSGA-II, which may indicate that IMOFA achieves a better balance between certain specific optimization objectives.

[0099] Table 4 compares the results of different algorithms in solving two cases of decommissioned engines. After merging and filtering all the better solutions, 12 Pareto-optimal solutions were obtained. As shown in Table 4, among the 12 Pareto solutions, IMOFA contributed 9 solutions, accounting for 75%; WCA contributed 2 solutions, accounting for 16.7%; and IWOA contributed 1 solution, accounting for 8.3%. This indicates that IMOFA has stronger optimization capabilities than the other four algorithms in terms of multi-objective solution sets.

[0100] Table 4 shows the results of different algorithms for solving two cases of decommissioned engines.

[0101]

[0102]

[0103] The four single-objective minimum decomposition schemes obtained by the IMOFA algorithm in Table 4 (i.e., items No. 1, 2, 3, and 5 under the IMOFA algorithm in Table 4) are respectively plotted into Gantt charts, as follows: Figures 7-10 As shown, each solution follows constraints such as task priority, direction, and disassembly mode.

[0104] As can be seen, the diversity of solutions provides companies with the flexibility to make decisions based on their specific objectives. For example, if a company prioritizes the number of workstations, it might choose... Figure 7 The corresponding No. 1 disassembly solution; if time efficiency is a consideration, then the following might be chosen. Figure 8 The corresponding disassembly solution No. 2; if profit is the primary consideration, then... Figure 10 The corresponding No.5 disassembly solution is more preferable, and for environmental and safety reasons, Figure 9 The corresponding No.3 disassembly solution would be the best choice.

[0105] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A hybrid linear localized destruction dismantling optimization method for recycling multiple decommissioned products, characterized in that: Includes the following steps: Step S1: Collect product disassembly information, and determine the objective function with disassembly efficiency, production line length, safety risks, and disassembly profit as optimization objectives. The objective function is as follows: The objective function shown in equation (1) represents the weighted sum minimization of the number of workstations in a workstation combination and the number of workstations; The objective function shown in equation (2) represents minimizing the smoothing exponent by reducing line length and ensuring a balanced idle time. The objective function shown in equation (3) indicates that dangerous tasks should be prioritized as much as possible, i.e., the time at which processing begins should be minimized. The objective function shown in equation (4) represents maximizing dismantling profits; The objective function satisfies the following constraints: The constraints in equations (8) and (9) indicate that all tasks need to be disassembled: The constraints in equations (10) and (11) indicate that each task needs to be assigned to its respective production line: The constraints in equation (12) indicate that some tasks are directionally restricted. Therefore, tasks with left-side directional restrictions can only be assigned to the left, while tasks with right-side directional restrictions can only be assigned to the right. Equations (13) and (14) indicate that not all tasks can be disassembled in a destructive manner; dangerous tasks and demanding tasks, due to their inherent properties, can only be disassembled non-destructively. Equation (15) indicates that, except for dangerous and demand-based tasks, other tasks can arbitrarily choose their disassembly mode; when the feasibility value of the destructive disassembly of the task is e i Greater than a defined threshold e max When r indicates that the task requires destructive disassembly, then r i =1, otherwise a disassembly mode will be randomly selected: Equations (16) and (17) indicate that tasks assigned to the same workstation must wait until their predecessor tasks are completed before they can be dismantled. Equation (18) indicates that each process must be completed within the cycle time of the specified workstation: Equations (19) and (20) indicate that in order to establish and Given the relationship that task i in production line h is assigned to the p side of the kth workstation before k, this constraint restricts task i and task s to be assigned to the same side of the same workstation: Equation (21) indicates that the disassembly of subsequent tasks can only begin after the preceding tasks assigned to the same-side workstations have been completed: Equation (22) indicates that the disassembly task can only begin after its preceding task has been completed: Equation (23) represents the time limit for tasks on public workstations: A·(1-y jik )+w hi ≥w (3-h)j +t (3-h)j ,i∈I h ,j∈I (3-h) ,k∈M (23) Equations (24) and (25) represent and y ijk The relationship between them: Equations (26) to (28) represent the upper and lower limits on the number of workstations: Equation (29) indicates that the workstations are turned on sequentially: Equations (30) to (33) represent the restrictions on opening the combined workstation: In the above formulas, m is the maximum number of available workstations; M is the set of workstation numbers, M = {1, 2, ..., m}; γ is the production line weighting coefficient, γ1 = 100, γ2 = 1; S k Enable the variable S for the combined workstation k =1 indicates that the k-th combined workstation is on, and S is off when the workstation is off. k =0; k is the workstation number index; p represents the position variable of the task assigned to the workstation, P is the side set, P = {p|1,2}, where p = 1 represents the left side, p = 2 represents the right side, p = h indicates that the side p of the current workstation corresponds to the production line h, p = 3-h indicates that the side of the current workstation is opposite to the production line; 3-h represents another production line corresponding to the current production line h, h = {1,2}, when h = 1, 3-h = 2, when h = 2, 3-h = 1; This indicates whether the p-side of the common workstation k between the two production lines is open; if open, then... otherwise h is the double-sided production line number, H is the set of double-sided production line numbers, H = {1, 2}; CT is the cycle time. This indicates whether the p-side of workstation k belonging to this production line in the common workstation is enabled for the h-th production line. If enabled, then... otherwise n h Let be the total number of disassembly tasks on the h-th production line; i,j,s are the disassembly task numbers. Assign variables to the task. This indicates that the i-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... t hi L represents the disassembly time of task i on production line h; k Indicates whether the public workstation is enabled, L k =1 indicates that it is turned on, L k =0 indicates that it is not enabled; h i This indicates that task i has a dangerous attribute, h i =1 indicates that task i is dangerous, h i =0 indicates that task i is not dangerous; G hi C represents the dismantling revenue that task i obtains on its production line; hi Let A be the cost of disassembling task i on its production line; A is a maximum real number. Enable variables for workstations on the h-th production line. If the k-th workstation on the h-th line is enabled, then... otherwise C h This represents the additional dismantling cost required to handle hazardous task i; r i Let r be the disassembly mode variable for task i. If task i is a destructive disassembly, then r i =1; otherwise r i =0; The disassembly time required for Task i to utilize destructive disassembly. The disassembly time required for task i to utilize non-destructive disassembly; h Let I be the set of dismantling task numbers for the h-th production line. h ={1,2,...,n h }; To utilize the dismantling benefits generated from destructive dismantling for task i; To utilize the dismantling benefits generated by non-destructive dismantling for task i; The dismantling costs incurred by utilizing destructive dismantling for task i; The disassembly cost incurred by utilizing non-destructive disassembly for task i; Assign variables to the task. The time indicates that the i-th task of the 3-h production line will be assigned to the p-side of the k-th workstation; otherwise... Assign variables to the task. When the time is specified, the i-th task of the h-th production line is assigned to the left side of the k-th workstation; otherwise... Assign variables to the task. When the i-th task of the 3-h production line is assigned to the left side of the k-th workstation, otherwise... T h1 T represents the set of disassembly task numbers on the left side of the workstation on the h-th production line. h2 This is the set of dismantling task numbers on the right side of the workstation on the h-th production line; Assign variables to the task. When the task is assigned, the i-th task of the h-th production line is assigned to the right side of the k-th workstation; otherwise... DS represents the set of tasks with demand; HS represents the set of tasks with hazardous attributes; e i e represents the feasibility value for destructive dismantling of task i; max To maximize the feasibility of dismantling, e max =0.5; To decompose the sequence variables for the task, This indicates that task i and task j are simultaneously assigned to the p-th side of combined workstation k on the h-th production line, with task i assigned before task j; otherwise, t hs The disassembly time of task s on the h-th production line; I (3-h) This is the set of dismantling task numbers for the other production line corresponding to the current production line h; t (3-h)s This indicates the disassembly time of task s on another production line corresponding to the current production line h; y sik To remove the sequence variable y from the public workstation sik =1 indicates that tasks i and j are assigned to the right side of production line h or the left side of production line 2 at workstation k, and not on the same side. Task i is assigned before task j. Otherwise, y sik =0; Assign variables to the task. This indicates that the j-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... Assign variables to the task. This indicates that task i and task j are simultaneously assigned to the p-th side of the combined workstation k, and task i is assigned before task j; otherwise... Assign variables to the task. This indicates that task i and task j are simultaneously assigned to the p-th side of the combined workstation k, and task j is assigned before task i; otherwise... Assign variables to the task. This indicates that the j-th task of the h-th production line is assigned to the p-side of the k-th workstation; otherwise, it is... w hi w represents the start time of task i on the h-th production line; hj t represents the start time of task j on the h-th production line; hj Let the disassembly time of task j on production line h be _j_. Let i represent the priority relationship between task i and task j in the h-th production line; Assign variables to the task, where 3-p represents the side opposite to the current side p. This indicates that the j-th task of the h-th production line is assigned to the opposite face of the p-th side of the k-th workstation; otherwise... Assign variables to the task. This indicates that the j-th task in the other production line number corresponding to the current production line h will be assigned to the p side of the k-th workstation; otherwise... y ijk y is the dismantling order variable for public workstations. ijk =1 indicates that tasks i and j are assigned to the right side of production line 1 or the left side of production line 2 of workstation k, respectively, and are not on the same side. Task i is assigned before task j. Otherwise, y ijk =0; y jik y is the dismantling order variable for public workstations. jik =1 indicates that tasks i and j are assigned to the right side of production line 1 or the left side of production line 2 of workstation k, respectively, and are not on the same side. Task j is assigned before task i. Otherwise, y jik =0; Let h be the switch variable for the production line. If one side of the h-th production line is turned on, then... otherwise This indicates whether the variable is enabled on the left side of the first production line; To indicate whether the variable on the right side of the second production line is enabled; c m This indicates the cost of starting up the workstation; Step S2: Initialize parameters and product information, generate an initial firefly population, perform Pareto screening on the initial firefly population to identify dominant individuals, and perform iterative calculations using the firefly algorithm. Step S3: Determine whether to restart the algorithm, and select different methods to filter and calculate non-dominated solutions based on the restart status of the algorithm; Step S4: Update the external file based on the non-dominated solution, and determine whether the maximum number of iterations has been reached. If not, return to step S3 to continue iterative calculation. If the maximum number of iterations has been reached, output the result in the external file.

2. The hybrid linear localized destructive dismantling optimization method for recycling multiple decommissioned products according to claim 1, characterized in that: The method for generating the initial firefly population is based on a real number encoding representation, where each set of feasible task disassembly sequences obtained by encoding is used as a firefly individual.

3. The hybrid linear localized destructive dismantling optimization method for recycling multiple decommissioned products according to claim 1, characterized in that: The method for determining whether to restart the algorithm in step S3 is as follows: record the non-dominated solution set of each generation. If the solution set remains unchanged for 5 generations, then restart the algorithm; otherwise, do not restart it.

4. The hybrid linear localized destructive dismantling optimization method for recycling multiple decommissioned products according to claim 1, characterized in that: If the algorithm does not restart, it selects each non-Pareto dominant individual in the initial population and performs a three-point crossover operation with a randomly selected current Pareto dominant individual. Then, it performs a mutation operation on the crossover individuals in the firefly population and calculates the non-dominated solution.

5. The hybrid linear localized destructive dismantling optimization method for recycling multiple decommissioned products according to claim 1, characterized in that: If the algorithm restarts, a new firefly population is randomly generated and the target value is calculated. The current firefly population is replaced with a randomly generated population of fireflies with higher brightness. Then, a mutation operation is performed to update the population. The new population is then mixed with the previous generation's non-dominated solution set and a non-dominated solution is calculated.

6. The hybrid linear localized destructive dismantling optimization method for recycling multiple decommissioned products according to claim 1, characterized in that: When the non-dominated solution N q If the number of solutions exceeds the number of non-dominated solutions N0 to be output, update the external file; otherwise, select the solution with N0 non-dominated solutions by ranking by crowding distance and use it to update the external file.

Images