Multi-workshop job task cooperative scheduling method based on tabu search and NSGA-II fusion algorithm

Abstract

The invention discloses a multi-workshop job task cooperative scheduling method based on tabu search and an NSGA-II fusion algorithm. The method comprises the following steps: S1, obtaining basic information of a job task set, a machine set and a workshop set in a workshop; s2, establishing a multi-workshop operation task collaborative scheduling multi-objective optimization model; s3, configuring a tabu search and NSGA-II fusion algorithm used for solving the multi-workshop job task collaborative scheduling multi-objective optimization model; s4, establishing an analytic hierarchy process decision model; s5, outputting a plan scheduling scheme; and step S6, carrying out visual display on the job task collaborative scheduling scheme. According to the method, the workpiece completion efficiency is improved, and the total machining cost and the total machine energy consumption are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of computer service technology and the field of industrial Internet, and relates to an optimization algorithm for solving the problem of collaborative scheduling of multi-shop operation tasks, and specifically to a collaborative scheduling method for multi-shop operation tasks based on taboo search and NSGA-Ⅱ fusion algorithm. Background Art

[0003] With the development of the industrial Internet, the production model of "Internet + Manufacturing" has also developed rapidly. In intelligent factories, advanced cutting-edge technologies such as computer technology and manufacturing technology are used to form a highly coordinated production system to achieve intelligent management and dynamic resource allocation. The coordination between modules ultimately achieves the goals of maximum production efficiency, minimum production cost, and minimum workshop energy consumption. At present, the production model is constantly developing in the direction of personalized consumer demand and diversified commodity supply. The number of process types has increased, the production process has become more complicated, and the production methods have become more diversified. Multi-variety small batches have gradually become the mainstream production model. Appropriate workshop scheduling solutions can greatly reduce the consumption of human, material and financial resources of enterprises, achieve the optimal allocation of resources, improve machine utilization and production efficiency, save energy and production costs, and are conducive to the sustainable development of enterprises.

[0004] Compared with the traditional product-based workshop layout, the process-based layout is more conducive to process management and has high equipment utilization, but the existing research work still has the following problems:

[0005] (1) Most of them consider the task scheduling problem from the perspective of a single product workshop, without considering issues such as collaborative optimization and resource sharing between tasks;

[0006] (2) The optimization objective faces problems such as singleness and incomplete calculation;

[0007] (3) Traditional algorithms are prone to fall into local search space.

[0008] Therefore, studying the multi-shop job scheduling problem can better optimize the industrial manufacturing process, improve production efficiency and product quality, and reduce production costs. It plays a vital role in the production capacity and economic benefits of the enterprise and is of great significance to the development of the manufacturing industry. Summary of the invention

[0009] In order to solve the above problems existing in the prior art, the present invention provides a method for collaborative scheduling of multi-shop tasks based on taboo search and NSGA-Ⅱ fusion algorithm for multi-shop task collaborative scheduling. The method establishes a multi-objective optimization model for collaborative scheduling of multi-shop tasks, proposes a taboo search and NSGA-Ⅱ fusion algorithm TSNSGA-Ⅱ solution, solves the problem by multi-chromosome real number coding of process sorting sequence, machine selection sequence, and workshop belonging sequence, improves the efficiency of workpiece completion, reduces the total processing cost, and reduces the total machine energy consumption.

[0010] The objective of the present invention is achieved through the following technical solutions:

[0011] A multi-shop job task collaborative scheduling method based on taboo search and NSGA-Ⅱ fusion algorithm includes the following steps:

[0012] Step S1, obtaining basic information of a job task set, a machine set and a workshop set in a workshop;

[0013] Step S2, establishing a multi-objective optimization model for collaborative scheduling of multi-shop operation tasks;

[0014] Step S3, configuring a taboo search and NSGA-Ⅱ fusion algorithm for solving a multi-objective optimization model for collaborative scheduling of multi-shop operation tasks;

[0015] Step S4, establishing a hierarchical analysis method decision model;

[0016] Step S5: Output the planned scheduling scheme;

[0017] Step S6: Visually display the collaborative scheduling plan for the job tasks.

[0018] Compared with the prior art, the present invention has the following advantages:

[0019] 1. The present invention takes the collaborative scheduling problem of multi-workshop operation tasks as the research object, establishes a multi-objective optimization mathematical model of minimizing the maximum completion time, minimizing the total processing cost, and minimizing the total processing energy consumption, realizes the information flow sharing among multiple workshops, and solves the problems of insufficient utilization of machine resources and single scheduling objectives.

[0020] 2. The present invention proposes a taboo search and fast non-dominated sorting genetic fusion algorithm TSNSGA-Ⅱ to solve the collaborative scheduling model of multi-shop operation tasks, and proposes a new mutation strategy based on taboo search, which uses a taboo table to taboo certain excellent solutions to avoid falling into the local optimal solution during the search process, thereby improving the local search capability of the algorithm.

[0021] 3. The present invention adopts the analytic hierarchy process (AHP) to balance three objectives and select the optimal processing scheduling scheme, eliminating the subjective bias and confusion of manual weighting, achieving a trade-off between multiple objectives and improving the scientific nature of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the collaborative scheduling method of multi-shop operation tasks based on the fusion algorithm of taboo search and NSGA-Ⅱ;

[0023] Figure 2 This is a flowchart for solving the problem using the fusion algorithm of taboo search and NSGA-Ⅱ;

[0024] Figure 3 It is a schematic diagram of the multi-chromosome encoding method;

[0025] Figure 4 This is a schematic diagram of the taboo search neighborhood structure;

[0026] Figure 5 It is the structure diagram of the AHP model;

[0027] Figure 6 Output graphs for planning and scheduling scenarios. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is further described below in conjunction with the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be included in the protection scope of the present invention.

[0029] The present invention provides a method for collaborative scheduling of multi-shop operation tasks based on a fusion algorithm of taboo search and NSGA-Ⅱ. The method establishes a multi-objective optimization model for the collaborative scheduling problem of multi-shop operation tasks, combines the elite selection strategy of NSGA-Ⅱ and the local search strategy of taboo search to design a fusion algorithm of taboo search and NSGA-Ⅱ TSNSGA-Ⅱ to solve the multi-objective optimization model, introduces the hierarchical analysis method, compares and weighs the three factors according to the actual needs of the factory, and comprehensively selects the optimal scheduling plan. Figure 1 As shown, the following steps are included:

[0030] Step S1: Obtain the task set J in the workshop = {J1, J2, ..., J n}、Workshop set L={L1,L2,…,L q} and the machine set M = {M1,M2,…,M m}, where n is the number of tasks, q is the number of workshops, and m is the number of machines. The specific steps are as follows:

[0031] Step S1.1: The set of workshop tasks is J = {J1, J2, ..., J n}, the subscript is the workpiece number, J in the set i represents the i-th workpiece, where i∈[1,n], there are n work tasks in total, and each workpiece may belong to a different category, that is, the processes involved are not exactly the same, and each workpiece is independent of each other;

[0032] Step S1.2: There are m machines in the factory processing this batch of production tasks, and the machine set is M = {M1, M2, ..., M m}, M in the set i represents the i-th machine, where i∈[1,m], and the machines are distributed in various workshops. The workshop set is L={L1,L2,…,L q}, L in the set i represents the i-th workshop, where i∈[1,q], there are q workshops in total, and the processing time of the same process on different machines is different.

[0033] Step S2: Establish a multi-objective optimization model for collaborative scheduling of multi-shop operation tasks. The specific steps are as follows:

[0034] Step S2.1, set the first objective function to minimize the maximum completion time T total The maximum completion time is the processing completion time CT of the last process in all workshops. ij , where i represents the serial number of the workpiece, there are n work tasks, i∈[1,n], j represents the serial number of the process, and workpiece i has JN i process, j∈[1,JN i ]. Since the transportation time in this scenario is less than its processing time, only the machine processing time is considered, and the first objective function f1 is shown in formula (1);

[0035]

[0036] Step S2.2, set the second objective function to minimize the total processing cost. The total processing cost includes the labor loading cost C corresponding to the machine lk0 and machine depreciation cost C lkd , and the transportation cost between workshops C tij , the depreciation cost takes into account the actual factory scenario and the commonly used calculation method based on fixed asset depreciation. In this problem, the workload method is used for calculation; in the raw material processing process, workers are required to be responsible for loading materials and switching parameters, so it is necessary to consider the labor costs that need to be paid to the workers; when the process needs to be processed in different workshops, it is necessary to consider the transportation cost between workshops, which is proportional to the transportation distance. The second objective function f2 is shown in formula (2).

[0037]

[0038] Among them, the number of workshops is q, l∈[1,q], and the number of machines in each workshop is MN l ,k∈[1,MN l ], C lk0 For machine M lk The corresponding labor cost, C lkd For machine M lk The depreciation cost is calculated based on the workload method commonly used for fixed asset depreciation. According to formula (3), C origlk is the original purchase price of the machine. Based on the data of manufacturing processing machines, the estimated net residual value rate of the machine is α lk is 5%, τ lk is the expected service life of the machine, ranging from [5,15], p ijlk The jth process of workpiece i on machine M lk Processing time on O ij represents the jth production process of the i-th workpiece; C tij It represents the transportation cost. When the workpiece is processed in the same workshop, no transportation cost is required. When the workpiece is not processed in the same workshop, the transportation cost is the transportation distance s. ij The product of cost and unit cost is shown in formula (4).

[0039] C lkd =C origlk (1-α lk ) / τ lk ·p ijlk (3)

[0040]

[0041] Step S2.3, set the third objective function to minimize the total machine energy consumption. The energy consumption of the machine includes the energy consumption during machine processing and the energy consumption when the machine is in standby mode. The third objective function f3 is shown in formula (5). The number of workshops is q, l∈[1,q], and the number of machines in each workshop is MN l ,k∈[1,MN l ], machine working energy consumption Q l The calculation is shown in formula (6), the machine standby energy consumption Q u The calculation is shown in formula (7). There are n tasks in total, i∈[1,n], j represents the sequence number of the process, and workpiece i has a total of JN i process, j∈[1,JN i ] LP lk Indicates machine M lk Load power, ULP lk Indicates machine M lkNo-load power, x ijlk is a decision variable, when process O ij Select on machine M lk When processing, its value is 1, otherwise it is 0, p ijlk The jth process of workpiece i on machine M lk Processing time on ET lk For machine M lk The processing end time.

[0042]

[0043] Step S2.4: Set the constraints followed by the multi-objective optimization model for collaborative scheduling of multi-shop operation tasks.

[0044] Processing tasks follow a certain process route;

[0045]

[0046] in, represents the actual processing sequence of workpiece i, O ij is the jth process of workpiece i, j∈[1,JN i ],R(J i ) is the process route that workpiece i needs to follow.

[0047] Each machine belongs to a workshop;

[0048]

[0049] Among them, X iw Indicates uncoded machine M i Whether it belongs to workshop w, if machine M i belongs to workshop w, then X iw is 1, otherwise X iw is 0.

[0050] Each workpiece is independent and there is no priority in the processing sequence;

[0051] S={J1,J2,…,J n} (10)

[0052] Among them, S is a set of workpieces, which contains n workpieces J i ,There is no priority in the processing order between workpieces, which conforms to the characteristics of the collection.

[0053] Each process is not allowed to be preempted during processing, that is, the difference between the end time and the start time is the processing time of the process on this machine;

[0054]

[0055] Among them, CT ij For process O ij End processing time, ST ij For process O ij The start processing time, x ijlk is a decision variable, when process O ij Select on machine M lk When processing, its value is 1, otherwise it is 0, p ijlk is the jth process O of workpiece i ij On machine M lk Processing time on.

[0056] When a process on the machine is not completed, other processes are not allowed to be processed;

[0057] ST i(j+1) -CT ij ≥0,i=1,2,…,n;j=1,2,…,JN i -1 (12)

[0058] Among them, ST i(j+1) is the j+1th process O of workpiece i ij+1 Start processing time, CT ij For process O ij The end processing time.

[0059] The same workpiece can only be processed in one process on one machine at the same time;

[0060]

[0061] Among them, x ijlk is a decision variable, when process O ij Select on machine M lk When processing, its value is 1, otherwise it is 0.

[0062] The maximum completion time is greater than or equal to the completion time of any workpiece;

[0063]

[0064] Among them, T total is the maximum completion time of the task processing, t i is the completion time of workpiece i. The maximum completion time should be less than or equal to the specified delivery date;

[0065] minT total ≤T Deadline (15)

[0066] Among them, T total is the maximum completion time of the task processing, T DeadlineThe factory's stipulated delivery date. The completion time of a workpiece is the end processing time of its last process;

[0067]

[0068] Among them, T i is the completion time of job i, is the finishing time of the last process of workpiece i.

[0069] The processing time of all operations is non-negative;

[0070]

[0071] Among them, p ijlk is the jth process O of workpiece i ij On machine M lk Processing time on.

[0072] Each machine can start working at time 0;

[0073]

[0074]

[0075] Among them, G(M lk ) is a binary variable. When the machine cannot start working, G(M lk ) is 0, otherwise it is 1. Each workpiece can be processed after the task is released;

[0076]

[0077]

[0078] Among them, H(J i ) is a binary variable, workpiece J i If processing cannot be started, it is 0; otherwise, it is 1.

[0079] There are precedence constraints between the processes of the same workpiece, but there are no precedence constraints between the processes of different workpieces;

[0080] Assume that machine preparation time and release time are not considered;

[0081] It is assumed that all processing machines have been inspected, are of good quality and have no faults.

[0082] Step S3: configure the taboo search and NSGA-Ⅱ fusion algorithm for solving the multi-objective optimization model of multi-shop job collaborative scheduling. The algorithm flow chart is as follows: Figure 2 As shown, the specific steps are as follows:

[0083] Step S3.1, set the algorithm input to be the processing workpiece information, processing machine information, and the processing time of each process on different machines, and the algorithm output is the process machine allocation plan P.

[0084] Step S3.2, problem encoding and decoding. The present invention adopts a multi-chromosome real number encoding method. The first chromosome is the process sequence OS, which is used to determine the processing order of all workpiece processes; the second chromosome is the machine selection sequence MS, which is used to determine the machine selected for all workpiece processes. It contains a workshop subsequence WS, which is used to determine which workshop the machine belongs to. The encoding length of all chromosomes is the sum of the total number of workpiece processing processes, such as Figure 3 shown.

[0085] When calculating the fitness value, the present invention uses a greedy left shift insertion method to decode the OS-MS-WS encoding scheme into the corresponding machine processing scheme. First, it is necessary to find the completion time of the previous process of the process. If the process is the first process, it does not need to be considered. Then find the completion time of the current processing task of the processing machine selected by the process. The maximum value of the two is the start time of this process.

[0086] Step S3.3, initialize the algorithm parameters. Set the population size popsize, the number of iterations generation, and the crossover probability P c , minimum mutation probability P mmin , maximum mutation probability P mmax , taboo table length T_length, taboo search times T_gen.

[0087] Step S3.4, population initialization. A hybrid strategy of random initialization and greedy initialization is used to generate the first generation parent population Population of size popsize f Random initialization means randomly selecting an idle machine for each process, and the greedy initialization strategy means that the process gives priority to the idle machine with short processing time when selecting the processing machine.

[0088] Step S3.5, calculate the fitness value corresponding to each individual. f The individuals in the process perform decoding operations and calculate the maximum completion time T, the total processing cost C, and the total machine energy consumption E.

[0089] Step S3.6: Perform fast non-dominated sorting stratification on the individuals in the population and calculate their crowding distance.

[0090] Step S3.7: Population from the parent population f A binary selection is performed in the process to select individuals W1 and W2, and process crossover and machine crossover are performed on individuals W1 and W2.

[0091] Step S3.8. Perform tabu search mutation on individuals W1 and W2, and the specific steps are as follows:

[0092] Step S3.8.1. Select the initial individual W1 or W2 as the current individual, and set the tabu list as an empty set;

[0093] Step S3.8.2. Calculate the fitness value of the current individual and add it to the tabu list;

[0094] Step S3.8.3. Search for neighborhood solutions according to the neighborhood rule, perform fast non-dominated sorting and crowding degree calculation on the fitness values of the neighborhood solutions, and select the optimal solution in the neighborhood according to the target priority. The neighborhood structure is as Figure 4 shown;

[0095] Step S3.8.4. Judge whether the optimal solution is in the tabu list. If the optimal solution is not in the tabu list, then this optimal solution is used as the current solution and added to the tabu list. Otherwise, it is necessary to judge according to the aspiration criterion whether the highest-level objective function value of this optimal solution is better than the objective value of the first solution in the tabu list. If so, this optimal solution is the current solution, the first solution in the tabu list is released from tabu, and the optimal solution is added to the tabu list; otherwise, select the sub-optimal solution in the neighborhood and assign it to the current solution and add it to the tabu list;

[0096] Step S3.8.5. If the number of iterations is less than the maximum number of iterations, then jump to Step S3.8.3. Otherwise, jump out of the loop and execute Step S3.9;

[0097] Step S3.9. Generate the offspring population Children f and merge the parent and offspring to generate a new population New_Population f .

[0098] Step S3.10. Generate a new parent population. Perform fast non-dominated sorting on the individuals in the new population New_Population f and calculate the crowding degree to obtain the Pareto solution set, and use the elitist retention strategy to obtain a new parent population Population f with a size of popsize.

[0099] Step S3.11. Iterative evolution. If the generation Gen < Generation, then jump to Step S3.7, Gen = Gen + 1. Otherwise, obtain the Pareto front solution set and end Algorithm S3.

[0100] Step S4. Establish an analytic hierarchy process decision-making model, and the specific steps are as follows:

[0101] Step S4.1. Establish a hierarchical structure model of the target layer - criterion layer - solution layer, such as Figure 5 shown.

[0102] Step S4.2: construct a judgment matrix for the entire criterion layer and the solution layer corresponding to a single criterion.

[0103] Step S4.3: Calculate the weight of each factor and the final score to make a decision at the solution level.

[0104] Step S5: Output the scheduling plan. Output the Gantt chart of the scheduling plan for the job tasks on the workshop machines, such as Figure 6 shown.

[0105] Step S6: Visualize the collaborative scheduling scheme for the job tasks. The specific steps are as follows:

[0106] Step S6.1, the client is used to input basic information of job tasks and machines in the workshop, and visually output the scheduling results of multi-workshop job tasks;

[0107] Step S6.2, a server is used to execute a computer program to implement the steps of a multi-shop job task collaborative scheduling method based on a taboo search and NSGA-Ⅱ fusion algorithm;

[0108] Step S6.3: Database end, used to store basic information of job tasks and workshop machines.

Claims

1. A multi-shop job task collaborative scheduling method based on taboo search and NSGA-Ⅱ fusion algorithm, characterized in that The method includes the following steps: Step S1: Obtain the basic information of the job task set, machine set, and workshop set in the workshop; Step S2: Establish a multi-objective optimization model for collaborative scheduling of multi-workshop job tasks; Step S3: Configure a tabu search and NSGA-II fusion algorithm for solving the multi-objective optimization model of collaborative scheduling of multi-workshop job tasks; Step S4: Establish an analytic hierarchy process decision-making model; Step S5: Output the planned scheduling plan; Step S6: Visualize the collaborative scheduling plan of job tasks.

2. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-Ⅱ fusion algorithm according to claim 1 is characterized in that The specific steps of step S1 are as follows: Step S1.1: The set of workshop tasks is J = {J1, J2, ..., J n }, the subscript is the workpiece number, J in the set i represents the i-th workpiece, where i∈[1,n], there are n work tasks in total, and each workpiece may belong to a different category, that is, the processes involved are not exactly the same, and each workpiece is independent of each other; Step S1.2: There are m machines in the factory processing this batch of production tasks, and the machine set is M = {M1, M2, ..., M m }, M in the set i represents the i-th machine, where i∈[1,m], and the machines are distributed in various workshops. The workshop set is L={L1,L2,…,L q }, L in the set i represents the i-th workshop, where i∈[1,q], there are q workshops in total, and the processing time of the same process on different machines is different.

3. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-Ⅱ fusion algorithm according to claim 1 is characterized in that The specific steps of step S2 are as follows: Step S2.1, set the first objective function to minimize the maximum completion time T total :The maximum completion time is the processing completion time CT of the last process in all workshops ij , where i represents the serial number of the workpiece, there are n tasks in total, i∈[1,n], j represents the serial number of the process, and workpiece i has JN i process, j∈[1,JN i ], the first objective function f1 is as follows: Step S2.2, set the second objective function to minimize the total processing cost: the total processing cost includes the manual loading cost C corresponding to the machine lk0 and machine depreciation cost C lkd , and the transportation cost between workshops C tij , the second objective function f2 is as follows: Among them, C lk0 For machine M lk The corresponding labor cost, C lkd For machine M lk The depreciation cost, C tij represents the transportation cost; the number of workshops is q, l∈[1,q], and the number of machines in each workshop is MN l ,k∈[1,MN l ]; Step S2.3: Set the third objective function as minimizing the total machine energy consumption: The energy consumption of the machine includes the energy consumption during machine processing and the energy consumption during machine standby. The third objective function f3 is shown as follows: Among them, Q l is the energy consumption of the machine, Q u The energy consumption of the machine in standby mode; Step S2.4: Set the constraint conditions followed by the multi-objective optimization model of collaborative scheduling of multi-workshop job tasks: The processing tasks follow a certain process route; in, represents the actual processing sequence of workpiece i, O ij is the jth process of workpiece i, j∈[1,JN i ],R(J i ) is the process route that workpiece i needs to follow; Each machine belongs to a workshop; Among them, X iw Indicates uncoded machine M i Whether it belongs to workshop w, if machine M i belongs to workshop w, then X iw is 1, otherwise X iw is 0; Each workpiece is independent, and there is no priority for the processing order; S={J1,J2,…,J n } Among them, S is a set of workpieces, which contains n workpieces J i ,There is no priority in the processing order between workpieces, which conforms to the characteristics of the set; Each process is not allowed to be preempted during processing, that is, the difference between the end time and the start time is the processing time of the process on this machine; Among them, CT ij For process O ij End processing time, ST ij For process O ij The start processing time, x ijlk is a decision variable, when process O ij Select on machine M lk When processing, its value is 1, otherwise it is 0, p ijlk is the jth process O of workpiece i ij On machine M lk Processing time on When a process of a certain process on the machine is not completed, the processes of other processes are not allowed to be processed; ST i(j+1) -CT ij ≥0,i=1,2,…,n;j=1,2,…,JN i -1 Among them, ST i(j+1) is the j+1th process O of workpiece i ij+1 Start processing time; The same workpiece can only process one process on one machine at the same time; Among them, when process O ij Select on machine M lk When processing, its value is 1, otherwise it is 0; The makespan is greater than or equal to the completion time of any workpiece; Among them, T i is the completion time of job i; The makespan should be less than or equal to the specified due date; as total ≤T Deadline Among them, T Deadline The stipulated delivery date for the factory; The completion time of a workpiece is the end processing time of its last process; in, is the finishing time of the last process of workpiece i. The processing time of all processes is non-negative; p ijlk ≥0 i=1,…,n;j=1,…,JN i ;l=1,…,q;k=1,…,MN l Each machine can start working at time 0; 1≤l≤q,1≤k≤MN l Among them, G(M lk ) is a binary variable. When the machine cannot start working, G(M lk ) is 0, otherwise it is 1; each workpiece can be processed after the task is released; Among them, H(J i ) is a binary variable, workpiece J i If processing cannot be started, it is 0, otherwise it is 1; There are precedence constraints between the processes of the same workpiece, and there are no precedence constraints between the processes of different workpieces; It is assumed that the machine setup time and release time are not considered; It is assumed that all processing machines have been tested, are in good quality, and have no faults.

4. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-II fusion algorithm according to claim 3 is characterized in that In step S2.2, C lkd Calculated according to the following formula: C lkd =C origlk (1-a lk ) / t lk ·p ijk Among them, C origlk is the original purchase price of the machine, α lk is the estimated net residual value of the machine, τ lk is the expected service life of the machine, ranging from [5,15], p ijlk The jth process of workpiece i on machine M lk Processing time on.

5. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-Ⅱ fusion algorithm according to claim 3 is characterized in that In step S2.2, when the workpiece is processed in the same workshop, no transportation cost is required; when the workpiece is not processed in the same workshop, the transportation cost is the transportation distance s. ij The product of the unit cost cost is shown in the following formula: Among them, O ij Represents the j-th production process of the i-th workpiece.

6. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-II fusion algorithm according to claim 3 is characterized in that In step S2.3, the machine working energy consumption Q l , Machine standby energy consumption Q u The calculation of is as follows: Among them, LP lk Indicates machine M lk Load power, ULP lk Indicates machine M lk No-load power, ET lk For machine M lk The processing end time.

7. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-Ⅱ fusion algorithm according to claim 1 is characterized in that The specific steps of step S3 are as follows: Step S3.1: Set the algorithm input as the workpiece processing information, processing machine information, and processing time of each process on different machines, and the algorithm output as the process-machine allocation plan P; Step S3.2: Problem encoding and decoding: Adopt a multi-chromosome real number encoding method. The first chromosome is the operation sequence OS, which is used to determine the processing order of all workpiece processes; the second chromosome is the machine selection sequence MS, which is used to determine the machines selected by all workpiece processes, and it includes a workshop belonging subsequence WS, which is used to determine which workshop the machine belongs to. The encoding length of all chromosomes is the sum of the total number of processing processes of each workpiece; Step S3.3, initialize algorithm parameters: set population size popsize, number of iterations generation, crossover probability P c , minimum mutation probability P mmin , maximum mutation probability P mmax , length of taboo table T_length, number of taboo searches T_gen; Step S3.4, population initialization: Use a mixed strategy of random initialization and greedy initialization to generate the first generation parent population Popultation of size popsize f ; Step S3.5, calculate the fitness value corresponding to each individual: Population f The individuals in the process perform decoding operations and calculate the maximum completion time T, the total processing cost C, and the total machine energy consumption E; Step S3.6: Perform fast non-dominated sorting and stratification on the individuals in the population, and calculate their crowding distances; Step S3.7: Population from the parent population f Perform binary selection, select individuals W1 and W2, and perform process crossover and machine crossover on individuals W1 and W2; Step S3.8: Perform tabu search mutation on individuals W1 and W2; Step S3.9: Generate the offspring population Children f , and merge the parent and child generations to generate a new population New_Population f ; Step S3.10: Generate a new parent population. f The individuals in the population are quickly non-dominated and the crowding degree is calculated to obtain the Pareto solution set. The elite retention strategy is used to obtain a new parent population of size popsize. f ; Step S3.11: Iterative evolution: If the generation Gen < Generation, then jump to step S3.7, Gen = Gen + 1, otherwise obtain the Pareto front solution set and end the S3 algorithm.

8. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-II fusion algorithm according to claim 7 is characterized in that The specific steps of step S3.8 are as follows: Step S3.8.1: Select the initial individual W1 or W2 as the current individual, and set the tabu list as an empty set; Step S3.8.2, calculate the fitness value of the current individual and add it to the taboo table; Step S3.8.3, search for neighborhood solutions according to the neighborhood rules, perform fast non-dominated sorting and congestion calculation on the fitness values ​​of the neighborhood solutions, and select the optimal solution in the neighborhood according to the target priority. The neighborhood structure is shown in Figure 4; Step S3.8.4, determine whether the optimal solution is in the taboo table. If the optimal solution is not in the taboo table, then this optimal solution is used as the current solution and added to the taboo table. Otherwise, it is necessary to determine whether the highest-level objective function value of the optimal solution is better than the objective value of the first solution in the taboo table according to the amnesty criteria. If so, then this optimal solution is the current solution, the first solution in the taboo table is unbanned, and the optimal solution is added to the taboo table; otherwise, the suboptimal solution in the neighborhood is selected and assigned to the current solution and added to the taboo table; Step S3.8.5: If the number of iterations is less than the maximum number of iterations, jump to step S3.8.3; otherwise, exit the loop and execute step S3.

9.

9. The method for collaborative scheduling of multi-shop operations based on taboo search and NSGA-Ⅱ fusion algorithm according to claim 1 is characterized in that The specific steps of step S4 are as follows: Step S4.1, establish a target layer-criteria layer-scheme layer hierarchical structure model; Step S4.2, constructing a judgment matrix for the entire criterion layer and the solution layer corresponding to a single criterion; Step S4.3: Calculate the weight of each factor and the final score to make a decision at the solution level.

10. The method for collaborative scheduling of multi-shop operations based on the fusion algorithm of taboo search and NSGA-Ⅱ according to claim 1 is characterized in that The specific steps of step S6 are as follows: Step S6.1, the client is used to input basic information of job tasks and machines in the workshop, and visually output the scheduling results of multi-workshop job tasks; Step S6.2, a server is used to execute a computer program to implement the steps of a multi-shop job task collaborative scheduling method based on a taboo search and NSGA-Ⅱ fusion algorithm; Step S6.3: Database is used to store basic information of job tasks and workshop machines.

Images