Multi-target self-adaptive collaborative production scheduling method, system and equipment considering collaborative transportation operation of multi-skill workers and logistics equipment, and medium

Abstract

The invention discloses a multi-target adaptive collaborative production scheduling method, system and equipment considering collaborative transportation operation of multi-skill workers and logistics equipment, and a medium. The method comprises the following steps: collecting production resource information, and constructing a multi-target collaborative scheduling model; processing by adopting a multi-layer coding structure to generate an initial scheduling population; performing fitness evaluation to obtain an initial evaluation result; executing evolutionary operation and local search to obtain an optimized scheduling population; screening to obtain an initial elite solution set; and presetting an iteration stop condition, repeatedly updating the initial elite solution set to obtain a final elite solution set, and outputting a final scheduling scheme from the final elite solution set. According to the method, the complex collaborative optimization problem of machine processing resources, logistics transportation resources and multi-skill worker resources is solved, the solving efficiency of a multi-target and strong-constraint scheduling problem is improved, a high-quality scheduling scheme can be obtained within reasonable time, the production efficiency and the resource utilization rate are improved, and the energy consumption and the cost are reduced.

Description

Technical Field

[0001] This invention relates to the field of production scheduling technology, and in particular to a multi-objective adaptive collaborative production scheduling method, system, equipment and medium that considers the collaborative transportation operations of multi-skilled workers and logistics equipment. Background Technology

[0002] Against the backdrop of industrial manufacturing moving towards intelligence and flexibility, production systems are becoming increasingly complex. Their efficient operation depends on the close collaboration of various resources. Among them, processing machines, automated guided vehicles (AGVs) and other logistics equipment, as well as skilled workers with multiple skills, are the three core resources that constitute a modern smart workshop. How to integrate and optimize machine processing sequences, AGV transportation paths and task allocation, and the scheduling of multi-skilled workers to achieve the overall optimization of production efficiency, resource utilization, and energy consumption has become a key challenge in the field of production scheduling.

[0003] However, existing production scheduling research and practice only focus on machine scheduling or the combination optimization of limited resources. They lack a complete model and method for the systematic coordinated scheduling of machine processing resources, logistics and transportation resources, and multi-skilled worker resources. This leads to poor connection between various resources in actual production, with each waiting for the other, which restricts the improvement of overall efficiency. In addition, in scenarios that consider the coordination of multiple resources, due to the variety of decision variables and the coupling and complexity of constraints, traditional scheduling algorithms often have defects such as low search efficiency, easy to get trapped in local optima, and difficulty in obtaining high-quality feasible solutions within a reasonable time when solving such large-scale, strongly coupled optimization problems. They cannot meet the requirements of real-time, dynamic, and multi-objective intelligent scheduling. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multi-objective adaptive collaborative production scheduling method, system, equipment, and medium that considers the collaborative transportation operations of multi-skilled workers and logistics equipment. This solves the problems of existing technologies that focus on single resource scheduling, lack collaborative scheduling, resulting in low collaborative efficiency. Furthermore, because collaborative scheduling for such problems is complex and highly coupled, traditional algorithms are difficult to solve, easily get trapped in local optima, and cannot meet the scheduling requirements of intelligent workshops.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment, including: Collect production resource information and construct a multi-objective collaborative scheduling model based on the production resource information; The multi-objective cooperative scheduling model is processed using a multi-layer coding structure to generate an initial scheduling population; The fitness of the initial scheduling population is evaluated using a fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation results of each initial scheduling solution in the initial scheduling population. Based on the initial evaluation results, evolutionary operations and local search are performed on the initial scheduling population to obtain an optimized scheduling population. Based on the initial evaluation results, the optimized scheduling population is selected to obtain an initial elite solution set; A preset iteration stopping condition is set, and the initial elite solution set is repeatedly updated until the iteration stopping condition is met to obtain the final elite solution set. The final scheduling scheme is then output from the final elite solution set.

[0007] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the steps of constructing the multi-objective collaborative scheduling model include: Based on the production resource information, determine the collaborative relationship between machine processing resources, logistics and transportation resources, and multi-skilled worker resources; Based on the aforementioned collaborative relationship, decision variables involving process sequence, machine processing speed, skilled worker allocation, logistics equipment allocation, and logistics equipment speed are defined. Based on the decision variables, construct a multi-objective function with at least two of the following as optimization objectives: minimizing completion time, minimizing total cost, minimizing resource idle time, and minimizing total energy consumption; Based on the collaborative relationship and the decision variables, construct constraints that include the machine processing process, the logistics and transportation process, the multi-skilled worker operation process, and the collaborative process; A multi-objective cooperative scheduling model is constructed based on the multi-objective function and the constraints.

[0008] The beneficial effects of this preferred technical solution are as follows: by collaboratively defining multiple types of decision variables, a multi-objective collaborative scheduling model containing multiple objective functions and constraints is constructed, which can comprehensively characterize and optimize the complex coupling relationship between machine processing resources, logistics and transportation resources and multi-skilled worker resources, thereby improving overall production efficiency and resource utilization.

[0009] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the step of generating the initial scheduling population includes: Based on the decision variables, a multi-layer coding structure is constructed, including a process sequence layer, a machine processing speed sequence layer, a skilled worker allocation sequence layer, a logistics equipment allocation sequence layer, and a logistics equipment speed sequence layer. Based on the multi-layer coding structure, multiple initial scheduling solutions that meet the constraints are generated, forming the initial scheduling population. The initial scheduling population is generated based on the multi-layer coding structure.

[0010] The beneficial effects of this preferred technical solution are as follows: by constructing a multi-layer coding structure that integrates the process sequence layer, machine processing speed sequence layer, skilled worker allocation sequence layer, logistics equipment allocation sequence layer, and logistics equipment speed sequence layer, it can represent all decision variables in the collaborative scheduling without omission, ensure that the generated initial scheduling solution strictly meets the production constraints, provide a high-quality and diverse initial scheduling population, and improve the convergence speed and global optimization capability of the algorithm.

[0011] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the step of obtaining the initial evaluation result of each initial scheduling solution in the initial scheduling population includes: Based on the initial scheduling solution, construct the comparison point sequence and the reference point sequence; The reference point sequence is converted into a reference fuzzy set, and the comparison point sequence is converted into a comparison fuzzy set; Calculate the correlation entropy coefficient between each of the reference fuzzy sets and the comparison fuzzy sets, and use it as the initial evaluation result of the corresponding initial scheduling solution.

[0012] The beneficial effects of this preferred technical solution are as follows: by introducing a fuzzy correlation entropy mechanism, the objective value of the initial scheduling solution is transformed into a comparable correlation entropy coefficient, which can adapt to the uncertainty in multi-objective collaborative optimization and improve the overall quality and convergence efficiency.

[0013] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the step of obtaining the optimized scheduling population includes: Based on the initial evaluation results and the multi-layer coding structure, crossover and mutation operations are performed on the initial scheduling population to obtain an intermediate scheduling population. Multiple local search strategies are preset, and the historical performance data of each local search strategy is determined based on the initial evaluation results. Based on the historical performance data, a target strategy is selected from the plurality of local search strategies; The target strategy is used to perform a local search on the intermediate scheduling population to obtain an optimized scheduling population.

[0014] The beneficial effects of this preferred technical solution are as follows: by combining global crossover operations, mutation operations, and local search, and selecting the optimal target strategy based on historical performance data, the global exploration capability and local development accuracy are enhanced, and the search efficiency during the optimization process is improved.

[0015] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the step of obtaining the initial elite solution set includes: Based on the correlation entropy coefficient in the initial evaluation results, the initial scheduling solutions in the optimized scheduling population are ranked to obtain candidate elite scheduling solutions. A maximum capacity is preset. Based on the sorting result and the maximum capacity, a corresponding number of candidate elite scheduling solutions are selected from the optimized scheduling population and added to the initially empty elite solution set to obtain the initial elite solution set.

[0016] The beneficial effects of this preferred technical solution are as follows: by sorting based on the correlation entropy coefficient and controlling the maximum capacity, the most representative candidate elite scheduling solutions can be selected from the optimized scheduling population in an efficient and accurate manner, forming an initial elite set, which lays the foundation for the output of the final scheduling scheme.

[0017] As a preferred embodiment of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment described in this invention, the steps of presetting an iteration stopping condition, repeatedly updating the initial elite solution set until the iteration stopping condition is met to obtain a final elite solution set, and outputting the final scheduling scheme from the final elite solution set include: A preset iteration stopping condition is set, the optimized scheduling population is used as the current generation scheduling population, and the initial elite solution set is used as the current elite solution set; Using the current generation of scheduling population as input, the steps of fitness evaluation, obtaining the optimal scheduling population, and obtaining the elite solution set are repeatedly executed. After each execution, the new generation of optimal scheduling population is updated to the current generation of scheduling population, and the new generation of elite solution set is updated to the current elite solution set. The iteration terminates when the iteration stopping condition is met, and the current elite solution set is taken as the final elite solution set, and the current elite scheduling solution in the current elite solution set is taken as the final elite scheduling solution. Select the final elite scheduling solution with the optimal correlation entropy coefficient from the final elite solution set, and output it as the final scheduling scheme.

[0018] The beneficial effects of this preferred technical solution are as follows: by iteratively updating the new generation of optimized scheduling population and the new generation of elite solution set, and outputting the optimal final elite scheduling solution after satisfying the preset iteration stopping condition, the final output scheduling scheme achieves comprehensive optimization on multiple objectives, balancing solution efficiency and optimization quality.

[0019] Secondly, the present invention provides a multi-objective adaptive collaborative production scheduling system that considers the collaborative transportation operations of multi-skilled workers and logistics equipment, comprising: The model building module is used to build a multi-objective collaborative scheduling model based on the production resource information. The population initialization module is used to process the multi-objective cooperative scheduling model using a multi-layer coding structure to generate an initial scheduling population; The evaluation module is used to evaluate the fitness of the initial scheduling population using a fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation results. The evolution and search module is used to perform evolutionary operations and local searches on the initial scheduling population based on the initial evaluation results to obtain an optimized scheduling population. The elite management module is used to select elite scheduling solutions from the optimal scheduling population based on the initial evaluation results and manage the updates of the elite solution set. An iteration control module is used to preset the iteration stop condition and control the repeated update process of the elite solution set until the iteration stop condition is met, thus obtaining the final elite solution set. The output module is used to output the final scheduling scheme from the final elite solution set.

[0020] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of a multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows: by constructing a complete multi-objective cooperative scheduling model, adopting a multi-layer coding structure and a fuzzy correlation entropy evaluation mechanism, and combining evolutionary operations and local search strategies, it solves the complex cooperative optimization problem of machine processing resources, logistics transportation resources and multi-skilled worker resources, improves the solution efficiency of multi-objective, strongly constrained scheduling problems, can obtain high-quality scheduling schemes within a reasonable time, improves production efficiency and resource utilization, and reduces energy consumption and costs. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the overall process of a multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment, as described in an embodiment of the present invention.

[0025] Figure 2 This diagram illustrates the collaborative scheduling problem of machine processing resources, AGV transportation resources, and multi-skilled worker resources.

[0026] Figure 3 A schematic diagram of five-layer collaborative coding.

[0027] Figure 4 This is a flowchart of the fuzzy correlation entropy evaluation mechanism.

[0028] Figure 5 This is a multi-strategy crossover method. Detailed Implementation

[0029] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0030] Example 1, referring to Figure 1 As an embodiment of the present invention, a multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment is provided, comprising: S100. Collect production resource information and construct a multi-objective collaborative scheduling model based on the production resource information.

[0031] S200. The multi-objective cooperative scheduling model is processed using a multi-layer coding structure to generate an initial scheduling population.

[0032] S300. The fitness of the initial scheduling population is evaluated using the fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation result of each initial scheduling solution in the initial scheduling population.

[0033] S400. Based on the initial evaluation results, perform evolutionary operations and local search on the initial scheduling population to obtain an optimized scheduling population.

[0034] S500. Based on the initial evaluation results, the optimized scheduling population is selected to obtain the initial elite solution set.

[0035] S600. Preset the iteration stop condition, repeatedly update the initial elite solution set until the iteration stop condition is met, obtain the final elite solution set, and output the final scheduling scheme from the final elite solution set.

[0036] It should be noted that traditional production scheduling methods focus on optimizing single or limited resources, lacking systematic and coordinated scheduling of machine processing resources, logistics and transportation resources, and multi-skilled worker resources. Due to the variety of decision variables and the complexity of constraint coupling, traditional optimization algorithms are inefficient and prone to getting trapped in local optima, making it difficult to support the real-time scheduling needs of multi-objective smart workshops.

[0037] Therefore, to address the aforementioned problems of difficulty in multi-resource coordination and low solution efficiency, the following steps (S100-S600) are employed: first, production resource information is collected and a multi-objective collaborative scheduling model is constructed to characterize the coupling relationships among multiple resources; then, an initial scheduling population is generated using a multi-layer coding structure, and the quality of the initial scheduling solution is evaluated using a fuzzy correlation entropy evaluation mechanism; the scheduling population is continuously optimized by combining evolutionary operations and local search, and high-quality initial scheduling solutions are retained through the initial elite solution set; finally, when the stopping condition is met, the comprehensive optimal final scheduling scheme is output, providing optimization decision support for achieving efficient, energy-saving, and low-cost collaborative production in intelligent workshops.

[0038] Example 2, refer to Figure 1 As an embodiment of the present invention, based on the above embodiment, a multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment is provided.

[0039] In this embodiment, S100 involves collecting production resource information and constructing a multi-objective collaborative scheduling model based on that information. Taking a smart manufacturing workshop as an application scenario, this workshop has multiple manufacturing units (MCs), multiple automated guided vehicles (AGVs), multiple skilled workers, a raw material warehouse, and a finished product warehouse. It needs to produce multiple workpieces with different process routes. Specifically, for A1~A5 in S100, the implementation is as follows: A1. Based on the production resource information, determine the collaborative relationship between machine processing resources, logistics and transportation resources, and multi-skilled worker resources.

[0040] Specifically, such as Figure 2As shown, the operations of machine processing resources, logistics and transportation resources, and multi-skilled worker resources must be sequentially connected in time and cannot overlap. The collaborative relationship is reflected in the following: each process of a workpiece must be completed by a skilled worker with the corresponding skills who has arrived at the designated manufacturing unit (MC) to operate the machine; before the process begins, the automated guided vehicle (AGV) needs to transport the workpiece from the location of the previous process or the raw material warehouse to the current MC and complete loading and other adjustments; after the process is completed, the AGV needs to transport the workpiece to the MC designated for the next process or the finished product warehouse; after completing a process, the skilled worker needs to transfer to the MC where the next process to be operated is located.

[0041] A2. Based on the aforementioned collaborative relationship, define decision variables involving process sequence, machine processing speed, skilled worker allocation, logistics equipment allocation, and logistics equipment speed.

[0042] Specifically, the decision variables mainly include the process ordering variable, which represents the processing sequence of each workpiece process on the machine; the machine processing speed selection variable, which assigns a specific processing speed level to each process; the skilled worker allocation variable, which assigns a multi-skilled worker capable of operating the corresponding machine to each process; the AGV allocation variable, which assigns an AGV to each transport task of each workpiece; and the AGV speed selection variable, which assigns a specific AGV transport speed level to each transport task.

[0043] A3. Based on the decision variables, construct a multi-objective function with at least two of the following as optimization objectives: minimizing completion time, minimizing total cost, minimizing resource idle time, and minimizing total energy consumption.

[0044] Specifically, the following four optimization objectives are defined, and at least two of them can be selected for collaborative optimization in actual optimization: Minimize maximum completion time That is, the completion time of the last process for all workpieces, calculated by the following formula: ,in, The total number of AGVs. For the first The end time of transportation for each AGV; Minimize total cost The main cost is skilled worker costs, calculated using the following formula: in, For the first The unit time cost of a skilled worker For workpiece The Each process is carried out by skilled workers In the machine Above speed Processing time For the corresponding 0-1 decision variables, and These are the distance and speed at which skilled workers can move between locations. To determine skilled workers Should the transfer process be executed? 0-1 decision variables; Minimize total idle time This includes the idle time of machines and AGVs, calculated using the following formula: in, For machines The completion time, For machines Adjustment time between adjacent workpieces , These represent the unloaded and loaded transport distances of the AGV, respectively. For the first Layered AGV transport speed, and For the corresponding 0-1 decision variables; Minimize total energy consumption Including energy consumption in machining Adjusting and preparing energy consumption machine standby power consumption AGV transportation energy consumption AGV standby power consumption The calculation formulas are as follows: in, , , , , These are respectively: machine processing power, preparation and adjustment power, machine standby power, AGV transportation power, and AGV standby power; Total energy consumption is ; To minimize the maximum completion time Minimize total cost Minimize total idle time and minimize total energy consumption To optimize the objective, a multi-objective function It can be represented as: A4. Based on the collaborative relationship and the decision variables, construct constraints that include the machine processing process, the logistics transportation process, the multi-skilled worker operation process, and the collaborative process.

[0045] Specifically, the constraints include: Machine processing constraints ensure the correct processing logic of operations on the machine. For example, operation priority constraints ensure that operations on the same workpiece are processed in the correct sequence. The formula is as follows: ,in, For workpiece The The start time of each process, For the machine The processing time; machine exclusivity constraint, the same machine can only process one operation at a time, the formula is: ; AGV transportation constraints are transportation task sequence constraints; an AGV can only perform one transportation task at a time, as shown in the formula: ,in, and The first The first AGV The end time of the first task and the first The start time of each task; Multi-skilled worker constraints manage worker task allocation, skill matching, and movement. For example, the skilled worker task sequence constraint ensures a skilled worker can only begin the next task after completing the previous one; the skilled worker exclusivity constraint states that a skilled worker can only handle one process at a time, which can be expressed as a set of decision variables. Determine whether the process is handled by skilled workers. In the machine Summation constraints for the upper processing: The process-skilled worker one-to-one constraint means that each process on a machine can only be handled by one multi-skilled worker with the corresponding skills, represented as... ; The coordination constraints among machine processing resources, logistics and transportation resources, and multi-skilled worker resources mean that the start time of a process must simultaneously meet three conditions: the workpiece is delivered to the AGV, the assigned skilled worker has arrived at the machine, and the machine's adjustment and preparation have been completed. The formula is: ,in, , , These represent the end time of the corresponding transportation task, the arrival time of the skilled worker, and the time when the machine is ready to be completed, respectively.

[0046] A5. Construct a multi-objective cooperative scheduling model based on the multi-objective function and the constraints.

[0047] Specifically, this involves defining all decision variables and establishing multiple objective functions. By integrating all the necessary constraints, a multi-objective cooperative scheduling model that can be mathematically optimized and solved is obtained.

[0048] In an optional implementation, step S100 may further assign different weights to different objective functions, wherein the steps are as follows: when constructing multi-objective functions, according to the actual production management focus, assign different weights to each objective function. Assign different weights The multi-objective problem is transformed into a weighted single-objective problem for preliminary solution or as a reference, and a weighted comprehensive objective is constructed. ,in .

[0049] In another optional implementation, step S100 may also introduce objectives such as maximizing resource utilization or throughput as needed. The steps are as follows: add objectives such as maximizing the average machine utilization to the objective function set to form a richer multi-objective optimization problem.

[0050] In this embodiment of the application, step S200 involves processing the multi-objective cooperative scheduling model using a multi-layer coding structure to generate an initial scheduling population. Step S200 includes steps B1 to B2: B1. Based on the decision variables, construct a multi-layer coding structure that includes a process sequence layer, a machine processing speed sequence layer, a skilled worker allocation sequence layer, a logistics equipment allocation sequence layer, and a logistics equipment speed sequence layer.

[0051] Specifically, the multi-layer coding structure is a vector. ,in: The process sequence sequence layer (OS) is a sequence based on the arrangement of workpieces. Each workpiece repeats according to the total number of its processes. For example, for a workpiece that needs to be produced in 2 stages... There are two processes. There are 3 processes, where OS represents the processing sequence of the processes. ; Both the machine processing speed sequence layer MV and the skilled worker allocation sequence layer WS are sequences of the same length as OS. In the machine processing speed sequence layer MV, each element is the processing speed level number selected for the corresponding process, such as 1, 2, ..., L. In the skilled worker allocation sequence layer WS, each element is the skilled worker number assigned to the corresponding process, such as 1, 2, ..., w, and the skilled worker must possess the skill to operate the machine assigned to that process. The logistics equipment allocation sequence layer AS is a sequence of lengths equal to the total number of transportation tasks. The sequence of equal lengths, where each element is the AGV number assigned to the corresponding transportation task, such as 1, 2, ..., a; the logistics equipment speed sequence layer AV is a sequence of equal lengths to AS, where each element is the AGV transportation speed level number selected for the corresponding transportation task, such as 1, 2, ..., R.

[0052] B2. Based on the multi-layer coding structure, generate multiple initial scheduling solutions that meet the constraints, forming the initial scheduling population. The initial scheduling population is generated based on the multi-layer coding structure.

[0053] Specifically, such as Figure 3 As shown, a process sequence OS is randomly generated. For each process in OS, a machine processing speed sequence MV is generated by randomly selecting a speed level from the set of available processing speed levels. For each process in OS, a skilled worker is randomly selected from the set of skilled workers with the skills to operate the available machines for that process, generating a skilled worker assignment sequence WS. For each transport task, one AGV is randomly selected from all available AGVs for assignment, generating a logistics equipment assignment sequence AS. For each transport task in AS, a logistics equipment speed sequence AV is generated by randomly selecting a speed level from all available transport speed levels, thus generating a complete coded individual as the initial scheduling solution. ; Independently generated through a random generation process Such a coded individual Each encoded individual, after decoding, must be verified to ensure it meets the constraints, thus forming a collection. The initial scheduling population of initial scheduling solutions.

[0054] In an optional implementation, step S200 may also incorporate a heuristic method based on priority rules, which involves generating a better sequence of process steps using rules such as the shortest processing time or the longest processing time, then combining this with random allocation to generate other coding layers, and adding them to the initial scheduling population.

[0055] In another optional implementation, the Latin hypercube sampling (LHS) method can also be used for uniform sampling in step S200. The steps are as follows: LHS sampling is performed on the values ​​of each coding layer, such as MV and WS, within their defined domain to form a sampling subset. Then, the sampling subset is combined with randomly generated OS, AS, etc., to generate an initial scheduling population.

[0056] In this embodiment of the application, step S300 involves using a fuzzy correlation entropy evaluation mechanism to evaluate the fitness of the initial scheduling population, thereby obtaining the initial evaluation result for each initial scheduling solution in the initial scheduling population. Step S300 includes C1~C3: C1. Based on the initial scheduling solution, construct the comparison point sequence and the reference point sequence.

[0057] Specifically, such as Figure 4 As shown, for each initial scheduling solution in the initial scheduling population , , To determine the initial scheduling population size, a vector of objective function values ​​is obtained by decoding the five-layer code and simulating its performance on a multi-objective function. ,in To optimize the number of objectives, such as ,correspond The objective function value vector is used as the sequence of comparison points for the initial scheduling solution; Calculate each optimization objective for the entire initial scheduling population. , , To optimize the total number of objectives, find the minimum value among all initial scheduling solutions. and maximum value Construct the minimum point sequence and the sequence of maximum points This serves as a reference point sequence for calculating the correlation entropy.

[0058] C2. Convert the reference point sequence into a reference fuzzy set, and convert the comparison point sequence into a comparison fuzzy set.

[0059] Specifically, for the minimum point sequence in the reference point sequence and each comparison point sequence Each of its optimization objectives The membership function maps to the interval [0, 1] as follows: in, This represents the membership degree value. It is a very small positive number, such as This is used to prevent the denominator from being zero, and to set the reference point sequence. Transform into a reference fuzzy set ,in Each comparison point sequence Transform into the corresponding comparison fuzzy set ,in .

[0060] C3. Calculate the correlation entropy coefficient between each of the reference fuzzy sets and the comparison fuzzy sets, and use it as the initial evaluation result of the corresponding initial scheduling solution.

[0061] Specifically, for each initial scheduling solution Calculate its comparison fuzzy set With reference fuzzy set Fuzzy correlation entropy coefficient between The calculation formula is as follows: Among them, the fuzzy correlation entropy coefficient The value ranges from 0 to 1; the larger the value, the more fuzzy the set. With reference fuzzy set The closer they are, the higher the similarity, and the better the initial scheduling solution. The better the overall performance, the higher the calculated value. As the initial scheduling solution The initial assessment results.

[0062] In an alternative implementation, step S300 may further include using a maximum point sequence to simplify calculations or accommodate different evaluation preferences. As a reference point sequence, the steps are: in C1, As a reference point sequence, the numerator in the membership function of C2 is changed to To maintain the rationality of the mapping.

[0063] In another optional implementation, different membership functions can be used for fuzzification in step S300, specifically in C2, where a Gaussian function is used. in The standard deviation is used to fuzzify the data before calculating the correlation entropy.

[0064] In this embodiment of the application, S400, based on the initial evaluation results, evolutionary operations and local search are performed on the initial scheduling population to obtain an optimized scheduling population. Step S400 includes D1~D4: D1. Based on the initial evaluation results and the multi-layer coding structure, perform crossover and mutation operations on the initial scheduling population to obtain the intermediate scheduling population.

[0065] Specifically, such as Figure 5 As shown, evolutionary operations include crossover and mutation. This embodiment uses the roulette wheel selection method, based on the correlation entropy coefficient of the initial evaluation results. Select two parent coding individuals from the initial scheduling population. and Multiple strategies are used to cross the five-layer coding, such as process sequence cross, machine speed and skilled worker allocation sequence cross, and AGV allocation and speed sequence cross. Two cutting points are randomly selected, and the gene values ​​of the middle part of the two parent process processing sequence layers OS are swapped to generate two offspring OS. The gene values ​​of the corresponding positions of the two parent machine processing speed sequence layers MV and skilled worker allocation sequence layers WS are swapped to generate offspring MV and offspring WS. The logistics equipment allocation sequence layer AS and logistics equipment speed sequence layer AV are divided into segments according to the workpiece. Two different workpiece segments are randomly selected and swapped to generate offspring AS and offspring AV. The newly generated five-layer code is combined into offspring code individuals. For the offspring encoded individuals obtained after crossover, mutation is performed with a small probability. Two positions in the coding layer are randomly selected to exchange gene values ​​or two artifact fragments, generating a set of offspring encoded individuals of the same size as the initial scheduling population, thus obtaining the intermediate scheduling population.

[0066] D2. Preset multiple local search strategies and determine the historical performance data of each local search strategy based on the initial evaluation results.

[0067] Specifically, a preset containing Local search pools for a local search strategy For the first generation of evolution, i.e., the operation on the initial scheduling population, the initial performance data of each local search strategy is calculated based on the initial evaluation results. A portion of offspring encoded individuals are randomly selected from the intermediate scheduling population, and for each offspring encoded individual... Apply local search strategy Obtain the new coded individual Calculate the local search strategy Performance improvement rate : in, For performance improvement rate, and The evaluation results are shown before and after applying the local search strategy. To prevent zero decimals, a statistical local search strategy is employed. The average value is used as historical performance data. .

[0068] D3. Based on the historical performance data, select the target strategy from the multiple local search strategies.

[0069] Specifically, based on the historical performance data of each local search strategy Calculate the probability of being selected : in, The probability of being selected. Given a small normal number, use a roulette wheel to randomly generate a random number between [0, 1]. Based on the probability of being selected Choose a local search strategy as the target strategy for this iteration. .

[0070] D4. Use the target strategy to perform a local search on the intermediate scheduling population to obtain an optimized scheduling population.

[0071] Specifically, the selected target strategy This process is applied to each child coded individual in the intermediate scheduling population. For each selected child coded individual, the operation specified by the target policy is executed, such as swapping two velocity sequence values, to generate a new candidate individual. The evaluation result of the candidate individual is calculated. If the candidate individual is better than the selected child coded individual, the selected child coded individual is replaced; otherwise, the selected child coded individual is retained. After completing the traversal process, the optimized scheduling population is obtained. .

[0072] In an optional implementation, step S400 may also employ a method different from the two-point intersection method, such as uniform intersection or sequential intersection. The steps are as follows: generate a binary mask of the same length as OS, determine from which parent the child inherits the corresponding operation from the binary mask, and repair any missing or repeated operations.

[0073] In another alternative implementation, step S400 may also involve a certain probability To randomly select a local search strategy, the steps are as follows: Generate a random number. ,like Then, a local search strategy is randomly selected from the local search pool as... Otherwise, choose using the roulette wheel selection method.

[0074] In this embodiment of the application, step S500 involves filtering the optimized scheduling population based on the initial evaluation results to obtain an initial elite solution set. Step S500 includes E1~E2: E1. Based on the correlation entropy coefficient in the initial evaluation results, the initial scheduling solutions in the optimized scheduling population are ranked according to their merits to obtain candidate elite scheduling solutions.

[0075] Specifically, for the obtained optimized scheduling population Each initial scheduling solution in Correlation entropy coefficient in the corresponding initial evaluation results According to the correlation entropy coefficient The values ​​are sorted in descending order to obtain the sorted result, which is the sorted scheduling solution sequence. in of The value is the largest. of The minimum value is the first in the scheduling solution sequence. indivual Marked as a candidate elite scheduling solution, The preset number of candidate elites is usually less than or equal to the initial scheduling population size. .

[0076] E2. Preset maximum capacity. Based on the sorting result and the maximum capacity, select a corresponding number of candidate elite scheduling solutions from the optimized scheduling population and add them to the initially empty elite solution set to obtain the initial elite solution set.

[0077] Specifically, a maximum capacity ArchiveSize is preset, and based on the E1 sorting results, the number of candidate elite scheduling solutions is determined. If the capacity is less than or equal to the maximum capacity ArchiveSize, then all of them will be stored. Each candidate elite scheduling solution is added to the initially empty elite solution set. If the number of candidate elite scheduling solutions... If the value is greater than ArchiveSize, then only the top ArchiveSize candidate elite solutions will be scheduled. Add to the initially empty elite solution set to obtain an initial elite solution set containing at most ArchiveSize solutions. .

[0078] In an optional implementation, step S500 may also consider the distribution in the target space. The steps are as follows: after E1 sorting, methods such as crowding degree calculation or cluster analysis are used to further screen out the best and most dispersed solutions from the top-ranked solutions as candidate elite scheduling solutions, and then the maximum capacity control of E2 is performed.

[0079] In another optional implementation, step S500 can also combine the construction of the initial elite solution set with the elite solution update strategy in subsequent iterations. The steps are as follows: In E2, not only based on the current sorting result, it is also considered whether the candidate elite scheduling solution is dominated. Only when the candidate elite scheduling solution is not dominated is it added to the elite solution set and the candidate elite scheduling solution dominated by it is removed.

[0080] In this embodiment, S600 involves repeatedly updating the initial elite solution set under a preset iteration stopping condition until the iteration stopping condition is met, thereby obtaining the final elite solution set, and outputting the final scheduling scheme from the final elite solution set. Step S600 includes F1~F4: F1. Preset the iteration stopping condition, take the optimized scheduling population as the current generation scheduling population, and take the initial elite solution set as the current elite solution set.

[0081] Specifically, two iteration stopping conditions are set, and the maximum number of generations to be sent is determined. With convergence threshold The resulting optimized scheduling population Set as the current generation scheduling population for the first generation iteration ,Right now The initial elite solution set obtained Set as the current elite solution set in the first generation iteration Right now Set the current iteration algebra counter. .

[0082] F2. Using the current generation of scheduling population as input, repeatedly execute the steps of fitness evaluation, obtaining the optimal scheduling population, and obtaining the elite solution set. After each execution, update the new generation of optimal scheduling population to the current generation of scheduling population, and update the new generation of elite solution set to the current elite solution set.

[0083] Specifically, for the first Iteration, Enter the iterative loop to schedule the current generation of the population. Each scheduling solution in Referring to steps C1-C3 of S300, a new reference point is constructed based on the target value range of the current generation's scheduling population, and its current evaluation result and correlation entropy coefficient are calculated. Referring to steps D1-D4 of S400, As a result of the current assessment, By performing evolutionary operations and local search, a new generation of optimally scheduled population is obtained. Referring to steps E1-E2 of S500, based on from Select elite scheduling solutions from the pool and compare them with the current elite solution set. Perform merging and filtering, such as retaining non-dominated solutions, and control the set size to not exceed the preset maximum capacity. To obtain a new generation of elite solutions ,Will Assign to As the current generation scheduling population for the next generation; Assign to As the current elite solution set for the next generation, the iterative algebra counter is updated to... .

[0084] F3. When the iteration stopping condition is met, the iteration is terminated, and the current elite solution set is taken as the final elite solution set. The current elite scheduling solution in the current elite solution set is taken as the final elite scheduling solution.

[0085] Specifically, after each iteration, it is determined whether the preset iteration stopping condition is met. The iteration stopping condition is: if the current iteration number is... Or the current elite solution set Compared to previous generations, such as the first five generations, the degree of improvement is as follows: The rate of change of the value is less than the convergence threshold. Then terminate the iteration loop and set the current elite solution set at the time of iteration termination. As the ultimate elite solution set Each current elite scheduling solution contained in the final elite solution set is a final elite scheduling solution.

[0086] F4. Select the final elite scheduling solution with the optimal correlation entropy coefficient from the final elite solution set and output it as the final scheduling scheme.

[0087] Specifically, for the ultimate elite solution set Compare the corresponding correlation entropy coefficients of all final elite scheduling solutions. Take the current evaluation result of the corresponding iteration algebra or re-evaluate using the final elite solution set, and select the final elite scheduling solution with the largest correlation entropy coefficient. ,Right now: Final Elite Dispatch The five layers of encoding are decoded to obtain the complete process sequence, the processing machine and speed of each process, the allocation of AGV and transportation speed, and the allocation scheme of skilled workers, which constitutes the executable final scheduling scheme and outputs the final scheduling scheme.

[0088] In an alternative implementation, step S600 may further include refining the current elite solution set in each iteration. Perform regular cleanup operations, the steps of which are: calculate every several generations. If the Euclidean distance between any two elite scheduling solutions is less than a set similarity threshold, the solution with the worse current evaluation result is removed to better maintain the distribution.

[0089] In another optional implementation, step S600 may further involve weighted scoring and reselection of solutions in the final elite solution set based on preference weights, the steps being: for each optimization objective... Assigning preference weights right Each initial scheduling solution Calculate the weighted target value ,in It is the normalized target value, selected The optimal one is output as the final scheduling scheme.

[0090] In summary, this invention constructs a multi-objective scheduling model based on the collaborative relationship between machine processing resources, logistics and transportation resources, and multi-skilled worker resources, and represents decision variables through a multi-layer coding structure; it obtains evaluation results using a fuzzy correlation entropy evaluation mechanism, and continuously optimizes the scheduling population by combining evolutionary operations and local search; and it outputs the final scheduling scheme through iterative updates, thereby achieving efficient collaborative optimization of multiple resources, improving production efficiency and resource utilization, reducing energy consumption and costs, and providing effective support for dynamic and refined production management in smart workshops.

[0091] Example 3 illustrates a schematic scheme for a multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment. It should be noted that the technical solution of this system considering multi-objective adaptive collaborative production scheduling of multi-skilled workers and logistics equipment is based on the same concept as the technical solution of the aforementioned multi-objective adaptive collaborative production scheduling method. Details not described in detail in this embodiment can be found in the description of the aforementioned multi-objective adaptive collaborative production scheduling method.

[0092] This embodiment also provides a multi-objective adaptive collaborative production scheduling system that considers the collaborative transportation operations of multi-skilled workers and logistics equipment, including: The model building module is used to build a multi-objective collaborative scheduling model based on the production resource information. The population initialization module is used to process the multi-objective cooperative scheduling model using a multi-layer coding structure to generate an initial scheduling population; The evaluation module is used to evaluate the fitness of the initial scheduling population using a fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation results. The evolution and search module is used to perform evolutionary operations and local searches on the initial scheduling population based on the initial evaluation results to obtain an optimized scheduling population. The elite management module is used to select elite scheduling solutions from the optimal scheduling population based on the initial evaluation results and manage the updates of the elite solution set. An iteration control module is used to preset the iteration stop condition and control the repeated update process of the elite solution set until the iteration stop condition is met, thus obtaining the final elite solution set. The output module is used to output the final scheduling scheme from the final elite solution set.

[0093] This embodiment also provides an electronic device suitable for multi-objective adaptive collaborative production scheduling considering the collaborative transportation operations of multi-skilled workers and logistics equipment, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as proposed in the above embodiment.

[0094] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-objective adaptive collaborative production scheduling method proposed in the above embodiments, which considers the collaborative transportation operations of multi-skilled workers and logistics equipment.

[0095] The storage medium proposed in this embodiment and the multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0096] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0097] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment, characterized in that, include: Collect production resource information and construct a multi-objective collaborative scheduling model based on the production resource information; The multi-objective cooperative scheduling model is processed using a multi-layer coding structure to generate an initial scheduling population; The fitness of the initial scheduling population is evaluated using a fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation results of each initial scheduling solution in the initial scheduling population. Based on the initial evaluation results, evolutionary operations and local search are performed on the initial scheduling population to obtain an optimized scheduling population. Based on the initial evaluation results, the optimized scheduling population is selected to obtain an initial elite solution set; A preset iteration stopping condition is set, and the initial elite solution set is repeatedly updated until the iteration stopping condition is met to obtain the final elite solution set. The final scheduling scheme is then output from the final elite solution set.

2. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 1, characterized in that, The steps to construct a multi-objective cooperative scheduling model include: Based on the production resource information, determine the collaborative relationship between machine processing resources, logistics and transportation resources, and multi-skilled worker resources; Based on the aforementioned collaborative relationship, decision variables involving process sequence, machine processing speed, skilled worker allocation, logistics equipment allocation, and logistics equipment speed are defined. Based on the decision variables, construct a multi-objective function with at least two of the following as optimization objectives: minimizing completion time, minimizing total cost, minimizing resource idle time, and minimizing total energy consumption; Based on the collaborative relationship and the decision variables, construct constraints that include the machine processing process, the logistics and transportation process, the multi-skilled worker operation process, and the collaborative process; A multi-objective cooperative scheduling model is constructed based on the multi-objective function and the constraints.

3. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 2, characterized in that, The steps for generating the initial scheduling population include: Based on the decision variables, a multi-layer coding structure is constructed, including a process sequence layer, a machine processing speed sequence layer, a skilled worker allocation sequence layer, a logistics equipment allocation sequence layer, and a logistics equipment speed sequence layer. Based on the multi-layer coding structure, multiple initial scheduling solutions that meet the constraints are generated, forming the initial scheduling population. The initial scheduling population is generated based on the multi-layer coding structure.

4. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 3, characterized in that, The steps to obtain the initial evaluation result for each initial scheduling solution in the initial scheduling population include: Based on the initial scheduling solution, construct the comparison point sequence and the reference point sequence; The reference point sequence is converted into a reference fuzzy set, and the comparison point sequence is converted into a comparison fuzzy set; Calculate the correlation entropy coefficient between each of the reference fuzzy sets and the comparison fuzzy sets, and use it as the initial evaluation result of the corresponding initial scheduling solution.

5. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 4, characterized in that, The steps to obtain an optimized scheduling population include: Based on the initial evaluation results and the multi-layer coding structure, crossover and mutation operations are performed on the initial scheduling population to obtain an intermediate scheduling population. Multiple local search strategies are preset, and the historical performance data of each local search strategy is determined based on the initial evaluation results. Based on the historical performance data, a target strategy is selected from the plurality of local search strategies; The target strategy is used to perform a local search on the intermediate scheduling population to obtain an optimized scheduling population.

6. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 5, characterized in that, The steps to obtain the initial elite solution set include: Based on the correlation entropy coefficient in the initial evaluation results, the initial scheduling solutions in the optimized scheduling population are ranked according to their merits to obtain candidate elite scheduling solutions and ranking results. A maximum capacity is preset. Based on the sorting result and the maximum capacity, a corresponding number of candidate elite scheduling solutions are selected from the optimized scheduling population and added to the initially empty elite solution set to obtain the initial elite solution set.

7. The multi-objective adaptive collaborative production scheduling method considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in claim 6, characterized in that, The steps of setting a preset iteration stopping condition, repeatedly updating the initial elite solution set until the iteration stopping condition is met to obtain the final elite solution set, and outputting the final scheduling scheme from the final elite solution set include: A preset iteration stopping condition is set, the optimized scheduling population is used as the current generation scheduling population, and the initial elite solution set is used as the current elite solution set; Using the current generation of scheduling population as input, the steps of fitness evaluation, obtaining the optimal scheduling population, and obtaining the elite solution set are repeatedly executed. After each execution, the new generation of optimal scheduling population is updated to the current generation of scheduling population, and the new generation of elite solution set is updated to the current elite solution set. The iteration terminates when the iteration stopping condition is met, and the current elite solution set is taken as the final elite solution set. The current elite scheduling solution in the current elite solution set is taken as the final elite scheduling solution. The final elite scheduling solution with the optimal correlation entropy coefficient is selected from the final elite solution set and output as the final scheduling scheme.

8. A multi-objective adaptive collaborative production scheduling system considering the collaborative transportation operations of multi-skilled workers and logistics equipment, employing the method described in any one of claims 1-7, characterized in that, include: The model building module is used to build a multi-objective collaborative scheduling model based on the production resource information. The population initialization module is used to process the multi-objective cooperative scheduling model using a multi-layer coding structure to generate an initial scheduling population; The evaluation module is used to evaluate the fitness of the initial scheduling population using a fuzzy correlation entropy evaluation mechanism to obtain the initial evaluation results. The evolution and search module is used to perform evolutionary operations and local searches on the initial scheduling population based on the initial evaluation results to obtain an optimized scheduling population. The elite management module is used to select elite scheduling solutions from the optimal scheduling population based on the initial evaluation results and manage the updates of the elite solution set. An iteration control module is used to preset the iteration stop condition and control the repeated update process of the elite solution set until the iteration stop condition is met, thus obtaining the final elite solution set. The output module is used to output the final scheduling scheme from the final elite solution set.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the multi-objective adaptive collaborative production scheduling method that considers the collaborative transportation operations of multi-skilled workers and logistics equipment as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the multi-objective adaptive collaborative production scheduling method for considering the collaborative transportation operations of multi-skilled workers and logistics equipment as described in any one of claims 1 to 7.

Images