Intelligent scheduling method, scheduling system, electronic equipment, medium and product

Abstract

The invention provides an intelligent scheduling method, a scheduling system, electronic equipment, a medium and a product, and relates to the field of intelligent manufacturing. The method comprises the following steps: acquiring production associated information of a plurality of production machines and job associated information of a plurality of jobs to be scheduled; the job association information comprises all processes corresponding to each to-be-scheduled job, randomly generating an initial scheduling scheme through a coding mode based on the production association information and the job association information, and performing iterative improvement on the initial scheduling scheme based on a scoring model and an optimization algorithm to obtain a target scheduling scheme. According to the scheme, improved iteration can be carried out on the generated initial scheduling scheme based on the optimization algorithm and the scoring model according to factory requirements to obtain a more preferable scheme, and the scheme can improve the production and operation efficiency when facing a large-scale task allocation scene, brings considerable economic benefits for a processing production line, and is suitable for popularization and application. And meanwhile, the scheme can also be suitable for complex trepanning working conditions, and the searching direction is simpler and more efficient.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent manufacturing, and in particular to an intelligent scheduling method, a scheduling system, an electronic device, a medium and a product. Background Art

[0002] In the manufacturing industry, production scheduling plays a vital role. It is about how to efficiently allocate resources to accurately meet production needs. However, production scheduling is complex and changeable, especially when faced with many realistic constraints such as multiple types of bottleneck material restrictions, diverse equipment switching times, complex process dependencies, variable order quantities, flexible equipment selection, and frequent equipment startup and shutdown. In such a complex and changeable production environment, how to effectively integrate and optimize these unique constraints to ensure that production scheduling decisions can prioritize the completion rate, while achieving load balancing and minimizing equipment switching time, thereby improving overall production efficiency and benefits is an urgent problem to be solved. Summary of the invention

[0003] The technical problem to be solved by the present disclosure is to overcome the defect in the prior art that it is impossible to ensure that production scheduling decisions can give priority to improving the order completion rate in a complex and changeable production environment, and to provide an intelligent scheduling method, scheduling system, electronic equipment, medium and product.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] The present disclosure provides an intelligent scheduling method, which includes:

[0006] Acquire production-related information of a plurality of production machines and job-related information of a plurality of jobs to be scheduled; the job-related information includes all processes corresponding to each of the jobs to be scheduled;

[0007] Based on the production-related information and the operation-related information, randomly generating an initial scheduling plan by coding;

[0008] The initial scheduling solution is iteratively improved based on the scoring model and the optimization algorithm to obtain a target scheduling solution.

[0009] In this solution, the generated initial scheduling solution can be improved and iterated based on the optimization algorithm and scoring model according to the factory needs to obtain a more preferred solution. When facing large-scale task allocation scenarios, this solution can improve production and operational efficiency and bring considerable economic benefits to the processing production line. At the same time, this solution can also be applied to complex nesting conditions, and the search direction is simpler and more efficient.

[0010] Preferably, the step of iteratively improving the initial scheduling solution based on the scoring model and the optimization algorithm to obtain the target scheduling solution includes:

[0011] Based on the optimization algorithm, the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process are improved to obtain an improved plan;

[0012] Scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain an improvement effect score and an initial scheduling effect score;

[0013] Determining whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score;

[0014] Repeat the above steps until a preset number of times is reached and / or a preset condition is met, and the final initial scheduling scheme is obtained as the target scheduling scheme.

[0015] In this solution, by continuously iterating and improving the order of process arrangements and machine allocation, the waiting time and machine idle time in the production process can be reduced, thereby improving overall production efficiency; the optimization algorithm can help allocate production resources more reasonably, and the scoring model can quantify the pros and cons of the scheduling solution, making the scheduling process more objective and accurate, which helps to improve the quality of the final scheduling solution.

[0016] Preferably, the step of improving the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process based on the optimization algorithm to obtain an improved plan includes:

[0017] Destroying the arrangement order of the processes in the initial scheduling scheme based on the target destruction operator to obtain a first scheme;

[0018] Reorder the processes in the first solution based on the target repair operator to obtain a second solution;

[0019] Based on preset rules, the production machine corresponding to each process in the second solution is changed to shorten the processing time of each process, thereby obtaining the improved solution.

[0020] In this solution, by selecting destruction operators and repair operators, optimizing the arrangement order of processes, and reallocating the production machines corresponding to each process based on preset rules, the process processing time can be reduced, thereby improving the overall production efficiency. The optimization algorithm can automatically select good operators to destroy and repair the solution, with a high probability of finding a better solution. At the same time, it can avoid the occurrence of local optimal solutions as much as possible. The production machines corresponding to the process are changed based on preset rules, making the scheduling plan more flexible and adaptable to changes in production needs.

[0021] Preferably, the determining whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score includes:

[0022] In response to the current improvement effect score being greater than the initial production scheduling effect score, iterating the initial scheduling solution;

[0023] In response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and the acceptance probability being greater than or equal to a preset probability, the improvement plan is replaced with the initial scheduling plan;

[0024] In response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and based on the acceptance probability being less than the preset probability, the initial scheduling plan is retained.

[0025] In this scheme, by iteratively comparing the initial scheduling effect score and the improvement effect score, the scheduling scheme can be continuously optimized until the best or satisfactory solution is found; the scheme can dynamically adjust whether to iterate based on the comparison result between the current improvement effect score and the initial scheduling effect score, thereby increasing the adaptability of the scheduling scheme; through the acceptance probability mechanism in the simulated annealing algorithm, even if the current improvement effect score is not better than the initial scheduling effect score, there is a certain probability of accepting a worse solution, thereby avoiding the algorithm from converging to a local optimal solution too early; the simulated annealing algorithm balances the algorithm's exploration, finding new and possibly better solutions, and using the currently found solutions by adjusting the temperature and acceptance probability, thereby increasing the possibility of finding a global optimal solution.

[0026] Preferably, the target destruction operator is at least one operator selected from a plurality of preset destruction operators in order of weight. If the weights are the same, the target destruction operator is at least one operator arbitrarily selected from the plurality of preset destruction operators.

[0027] The target repair operator is at least one operator selected from a plurality of preset repair operators in order of weight. If the weights are the same, the target repair operator is at least one operator arbitrarily selected from the plurality of preset repair operators.

[0028] After the step of iterating the initial scheduling scheme, the method further includes updating the weights of the target destruction operator and the target repair operator corresponding to the current improvement effect score.

[0029] In this scheme, by selecting the destruction operator and the repair operator, the algorithm can adaptively select the most effective operator for operation, thereby enhancing the flexibility and adaptability of the algorithm; by updating the weights of the target destruction operator and the target repair operator during the iteration process, the algorithm can dynamically adjust the frequency and intensity of operator use according to the current optimization effect, thereby further improving the optimization efficiency; by iteratively updating the operator weights and selecting operators with larger weights, the algorithm can better escape from the local optimal solution and increase the probability of finding the global optimal solution.

[0030] Preferably, the scoring model includes at least one preset scoring condition, and the step of scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain the improvement effect score and the initial scheduling effect score includes:

[0031] Scoring the improvement plan according to each of the preset scoring conditions to obtain an improvement scoring result corresponding to each of the preset scoring conditions;

[0032] Accumulating each of the improvement scoring results to obtain the improvement effect score;

[0033] Scoring the initial scheduling scheme according to each of the preset scoring conditions to obtain an initial scoring result corresponding to each of the preset scoring conditions;

[0034] Each of the initial scoring results is accumulated to obtain the initial production scheduling effect score.

[0035] In this scheme, the improvement scheme and the initial scheduling scheme are scored through at least one preset scoring condition, which can comprehensively evaluate all aspects of the scheme and ensure the comprehensiveness and meticulousness of the scheme optimization; the improvement scoring results are accumulated to obtain the improvement effect score, and the initial scoring results are accumulated to obtain the initial scheduling effect score, so that the improvement effect can be quantified for easy comparison and analysis.

[0036] Preferably, the preset scoring conditions include a delivery rate scoring condition, a balanced production scoring condition and a production changeover time scoring condition.

[0037] In this solution, by optimizing the scheduling plan to improve the delivery rate, we can ensure that customer orders are completed on time, thereby improving customer satisfaction and loyalty; balanced production can ensure that there will be no bottlenecks in each link of the production line and improve the overall resource utilization; optimizing the conversion time can improve the production line's ability to adapt to changes in demand for different products and increase the flexibility of the production line.

[0038] The present disclosure also provides an intelligent scheduling system, the intelligent scheduling system comprising:

[0039] A correlation information acquisition module, used to acquire production correlation information of multiple production machines and job correlation information of multiple jobs to be scheduled; the job correlation information includes all processes corresponding to each of the jobs to be scheduled;

[0040] An initial scheduling scheme generating module, used to randomly generate an initial scheduling scheme by coding based on the production-related information and the operation-related information;

[0041] The target scheduling scheme determination module is used to iteratively improve the initial scheduling scheme based on the scoring model and the optimization algorithm to obtain the target scheduling scheme.

[0042] Preferably, the target scheduling solution determination module includes an improved solution acquisition unit, a score acquisition unit, an iteration determination unit and a target scheduling solution determination unit;

[0043] The improvement plan acquisition unit is used to improve the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process based on the optimization algorithm to obtain an improvement plan;

[0044] The scoring acquisition unit is used to score the improvement plan and the initial scheduling plan based on the scoring model to obtain an improvement effect score and an initial scheduling effect score;

[0045] The iteration determination unit is used to determine whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score;

[0046] The target scheduling scheme determining unit is used to repeat the above steps until a preset number of times is reached and / or a preset condition is satisfied, and the final initial scheduling scheme is obtained as the target scheduling scheme.

[0047] Preferably, the improved solution acquisition unit includes a first solution acquisition subunit, a second solution acquisition subunit, and an improved solution acquisition subunit;

[0048] The first scheme acquisition subunit is used to destroy the arrangement order of the processes in the initial scheduling scheme based on the target destruction operator to obtain the first scheme;

[0049] The second solution acquisition subunit is used to reorder the processes in the first solution based on the target repair operator to obtain a second solution;

[0050] The improvement scheme acquisition subunit is used to change the production machine corresponding to each process in the second scheme based on preset rules to minimize the processing time of each process and obtain the improvement scheme.

[0051] Preferably, the iterative determination unit is specifically used to iterate the initial scheduling scheme in response to the current improvement effect score being greater than the initial production scheduling effect score;

[0052] The iterative determination unit is further configured to replace the improvement plan with the initial scheduling plan in response to the current improvement effect score being less than or equal to the initial scheduling effect score, and obtaining an acceptance probability based on a simulated annealing algorithm, and the acceptance probability being greater than or equal to a preset probability;

[0053] The iterative determination unit is specifically used to respond to the current improvement effect score being less than or equal to the initial scheduling effect score, and obtaining the acceptance probability based on the simulated annealing algorithm, and retaining the initial scheduling plan based on the acceptance probability being less than the preset probability.

[0054] Preferably, the target destruction operator is at least one operator selected from a plurality of preset destruction operators in order of weight size. If the weight sizes are the same, the target destruction operator is at least one operator arbitrarily selected from the plurality of preset destruction operators. The target repair operator is at least one operator selected from a plurality of preset repair operators in order of weight size. If the weight sizes are the same, the target repair operator is at least one operator arbitrarily selected from the plurality of preset repair operators.

[0055] The target scheduling scheme determination module also includes an operator weight updating unit;

[0056] The operator weight updating unit is used to update the weights of the target destruction operator and the target repair operator corresponding to the current improvement effect score.

[0057] Preferably, the scoring model includes at least one preset scoring condition, and the scoring acquisition unit includes an improved scoring result acquisition subunit, an improved effect score acquisition subunit, an initial scoring result acquisition subunit and an initial production scheduling effect score acquisition subunit;

[0058] The improvement scoring result acquisition subunit is used to score the improvement plan according to each of the preset scoring conditions to obtain an improvement scoring result corresponding to each of the preset scoring conditions;

[0059] The improvement effect score obtaining subunit is used to accumulate each of the improvement scoring results to obtain the improvement effect score;

[0060] The initial scoring result acquisition subunit is used to score the initial scheduling scheme according to each of the preset scoring conditions to obtain an initial scoring result corresponding to each of the preset scoring conditions;

[0061] The initial production scheduling effect score obtaining subunit is used to accumulate each of the initial scoring results to obtain the initial production scheduling effect score.

[0062] The present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and used to run on the processor, and the processor implements the above-mentioned intelligent scheduling method when executing the computer program.

[0063] The present disclosure also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the intelligent scheduling method as described above is implemented.

[0064] The present disclosure also provides a computer program product, including a computer program, which implements the above-mentioned intelligent scheduling method when executed by a processor.

[0065] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0066] The positive and progressive effects of this disclosure are:

[0067] The solution disclosed in the present invention can improve and iterate the generated initial scheduling solution based on the optimization algorithm and scoring model according to the factory needs to obtain a more preferred solution. When facing large-scale task allocation scenarios, this solution can improve production and operational efficiency and bring considerable economic benefits to the processing production line. At the same time, this solution can also be applied to complex nesting conditions, and the search direction is more concise and efficient.

[0068] This solution can comprehensively consider resource constraints, switching time and process dependencies, quickly evaluate multiple solutions on the premise of meeting the constraints, and solve the optimal production scheduling plan; for production scheduling problems in complex situations, it integrates multi-dimensional evaluation indicators, stability and timeliness, and ease of operation to establish a complete solution method, solve the problem with the goals of high order completion rate, load balancing and short switching time, output the optimal production scheduling plan, and realize intelligent production. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A flowchart of an intelligent scheduling method provided in Embodiment 1 of the present disclosure;

[0070] Figure 2 This is a schematic diagram of the interaction between an encoder, a decoder, and an optimization algorithm provided in Embodiment 1 of the present disclosure;

[0071] Figure 3 A schematic diagram of a flow chart of an optimization algorithm provided in Embodiment 1 of the present disclosure;

[0072] Figure 4This is a module schematic diagram of an intelligent scheduling system provided in Embodiment 2 of the present disclosure;

[0073] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Example 3 of the present disclosure. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0075] It should be noted that if this embodiment involves directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0076] In addition, if there are descriptions involving "first", "second", etc. in this embodiment, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present disclosure.

[0077] In the embodiments of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0078] Example 1

[0079] See also Figures 1 to 3 , this embodiment provides an intelligent scheduling method, Figure 1This is a flowchart of an intelligent scheduling method provided in Example 1 of the present disclosure. This specification provides method operation steps such as the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the actual system or server product is executed, it can be executed in the order shown in the embodiment or the figure or in parallel (for example, in a parallel processor or multi-threaded processing environment). Specifically, Figure 1 As shown, the intelligent scheduling method includes:

[0080] S101: Acquire production-related information of a plurality of production machines and job-related information of a plurality of jobs to be scheduled; the job-related information includes all processes corresponding to each job to be scheduled.

[0081] In this embodiment, in a job shop environment, the production-related information of the production machine is the equipment number of the production machine, and the equipment numbers of all existing equipment are obtained and recorded; the job-related information of the job to be scheduled includes the order number of each job to be scheduled. If the job to be scheduled consists of multiple processes, all processes corresponding to each job to be scheduled also need to be recorded. Several jobs J = {J1, J2, ..., Jm} are scheduled on a group of machines M = {M1, M2, ..., Mm}. For example, there are 3 production machines, namely machine M1, machine M2 and machine M3, and there are also three jobs to be scheduled, namely J1, J2, and J3, where each job consists of multiple operations, that is, multiple processes. Taking J1 as an example, which contains three processes, it will appear three times in the scheduling process. The first J1 that appears in order is the first process, and the second J1 that appears is the second process, and so on. It should be noted that the current problem does not involve order splitting.

[0082] S102: Based on the production-related information and the operation-related information, an initial scheduling plan is randomly generated by coding.

[0083] Figure 2 This is a schematic diagram of the interaction between an encoder, a decoder and an optimization algorithm provided in Embodiment 1 of the present disclosure, such as Figure 2As shown, in this embodiment, the encoder first randomly generates an initial scheduling plan by encoding based on production-related information and job-related information. For example, assume that there is a small factory with 3 machines M = {M1, M2, M3} and 4 jobs J = {J1, J2, J3, J4}. Each job contains different processes, as follows: J1: process 1-1, process 1-2; J2: process 2-1; J3: process 3-1, process 3-2; J4: process 4-1; The encoder first receives the information of the above machines and jobs, including the processes included in each job. The encoder randomly generates an initial solution, for example: M1 machine will process: process 1-1, process 3-1; M2 machine will process: process 2-1, process 4-1; M3 machine will process: process 1-2, process 3-2; This initial solution is randomly generated and may not be perfect, but it provides a starting point for the optimization algorithm.

[0084] The jobs to be scheduled involve non-bottleneck material orders and may also involve bottleneck material orders. In the process of randomly generating the initial scheduling plan by coding, the non-bottleneck material orders are processed, that is, the Figure 2 In other words, the pre-constraint is that the job to be scheduled is a non-bottleneck material order. If there is no bottleneck material order, the initial scheduling plan can be directly generated by the encoder; if a bottleneck material order is involved, the bottleneck material order is not processed by the encoder to generate the initial scheduling plan, but is directly allocated to the production machine by the subsequent decoder before the machine processes the initial scheduling plan. The decoder processes the bottleneck material order as follows:

[0085] All orders are inserted according to the number of bottleneck material orders per day, and the remaining bottleneck material orders are retained until the next day. Equipment selection takes into account equipment load balancing to keep the processing end time as consistent as possible and the shortest processing time. in Indicates the order quantity, It indicates the processing time of the current device for each order of the current task. The load balancing weight of the device and the weight of the shortest processing time can be adjusted as needed, and so on.

[0086] The decoder also needs to decode the initial scheduling scheme generated by the encoder, that is, it can decode each group of M and J, arrange each process on each device in the order arranged in the initial scheduling scheme, and consider all constraints and perform score calculations. The specific steps are as follows:

[0087] Equipment order process insertion: According to each group of machines M and operations J, first according to Arrangement of tasks, including Indicates the end time of the previous process of the current order, if there is none, it is 0. Indicates the end time of the previous task of the current device; then the idle time of the current device is calculated to move the task forward. It is worth noting that in order to prevent different devices, different tasks, and different switching times from affecting the current operation, a maximum switching time is added to each task. After all tasks are assigned, the accurate switching time is updated, and the process dependency is considered to eliminate the excess switching time. At the same time, the information of each order of each device is counted.

[0088] After arranging the processes on the machines, it is necessary to score the solution. The decoder has a preset scoring model, which considers the production process of n processes on m machines. 1 = Number of orders delivered on time / total number of orders, balanced production score Production conversion time score A 3 =1-total switching time / (total production time of all equipment / 2), Note: (1) On-time delivery: the date with the maximum completion time of all work orders in the order ≤ the order requirement date; (2) The completion time of equipment j is equal to the end time of the last work order of the equipment minus the start time of the first work order; (3) The total production time of equipment j is equal to the end time of all work orders on the equipment minus the sum of the start time; Therefore, the scoring model calculates the score of the solution: A = 0.5 × delivery rate score A 1 +0.3×Balanced Production Rating A 2 +0.2×Transfer Time Score A 3 0.5, 0.3 and 0.2 are the weights of the delivery rate score, balanced production score and conversion time score respectively. The weights can be modified according to actual needs.

[0090] S103: Iteratively improve the initial scheduling plan based on the scoring model and optimization algorithm to obtain a target scheduling plan.

[0091] In this embodiment, the steps of iteratively improving the initial scheduling solution based on the scoring model and the optimization algorithm to obtain the target scheduling solution include:

[0092] S1031: improving the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process based on the optimization algorithm to obtain an improved plan;

[0093] In this embodiment, the order of the processes is optimized. The optimization algorithm adopts an adaptive large neighborhood search algorithm, and a breaking operator and a repairing operator are designed. In each iteration, the neighborhood of the current solution is obtained by the operator to search for a new feasible solution.

[0094] S1032: Scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain an improvement effect score and an initial scheduling effect score;

[0095] S1033: Determine whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score;

[0096] S1034: Repeat the above steps until a preset number of times is reached and / or a preset condition is satisfied, and the final initial scheduling solution is obtained as the target scheduling solution.

[0097] In this solution, by continuously iterating and improving the order of process arrangements and machine allocation, the waiting time and machine idle time in the production process can be reduced, thereby improving overall production efficiency; the optimization algorithm can help allocate production resources more reasonably, and the scoring model can quantify the pros and cons of the scheduling solution, making the scheduling process more objective and accurate, which helps to improve the quality of the final scheduling solution.

[0098] Specifically, the steps of improving the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process based on the optimization algorithm to obtain the improved plan include:

[0099] S10311, destroying the arrangement order of the processes in the initial scheduling plan based on the target destruction operator to obtain a first plan;

[0100] In this embodiment, the destruction operator design uses a larger destruction range in the early stage of the search to facilitate better search, and a smaller destruction range in the later stage to facilitate convergence. The initial destruction coefficient Q is designed 0 and the final damage coefficient Q 1 , the destruction coefficient of the current iteration round Where i is the current iteration number, f is the maximum iteration number, the maximum iteration number can be preset according to the actual situation, and the number of destructions in the current iteration round is designed to be Where N is the total number of processes, Indicates the rounding up of x. When destroying, only the order of each process is considered, and the order of the undestroyed processes remains unchanged. According to the characteristics of the problem, the destruction operator is designed, including the random destruction operator, that is, randomly removing a given number of processes, the maximum switching time operator, that is, removing the given number of pairs of the previous and next processes with the maximum switching time, and the maximum waiting time operator, that is, removing the given number of pairs of the previous and next processes with the maximum waiting time.

[0101] Specifically, the destruction operator is a preset operator, and there are multiple preset destruction operators. In the specific process of process destruction, at least one operator can be selected from multiple preset destruction operators in order of weight to destroy the process. If the weights are the same, at least one operator can be arbitrarily selected from multiple preset destruction operators to destroy the process. For example, one destruction operator can destroy one of the processes. For example, destruction operator A destroys process 1066, and destruction operator B destroys process 1067. If process 1066 and process 1067 need to be destroyed at the same time, destruction operator A and destruction operator B need to be used at the same time. At the beginning, the weight of each destruction operator is the same, and the initial destruction coefficient Q 0 and the final damage coefficient Q 1 Parameter design is based on experience. Assuming there are 100 processes in total, for example, we start to optimize 10% of the total processes, that is, the initial damage coefficient Q 0 , that is to say, the destruction operator destroys 100*10%=10 of 1066, 1067, 1066, 1066, 1067…, that is, Here, 1066 and 1067 refer to the jobs to be scheduled. The first occurrence of 1066 is the first process of the job to be scheduled 1066, the second occurrence is the second process of the job to be scheduled 1066, the third occurrence is the third process of the job to be scheduled 1066, and so on. The final iteration stops at 5% of the total optimization process, that is, the final damage coefficient Q 1 , it is getting smaller and smaller because the solution is generally getting better, and we need to reduce the search space to make the algorithm converge, or in other words, most of them are already optimal and only a small part needs to be adjusted. This formula is designed to allow the ratio to decrease smoothly according to the number of iterations, for example, from 10% to 5%.

[0102] Assume that the code is 1066 processed on device 001, 1067 processed on device 001, 1066 processed on device 001, 1066 processed on device 001, and 1067 processed on device 002. The same device, such as 001, processes different jobs to be scheduled, that is, there will be a switching time from order 1066 to 1067. If the maximum is 5 minutes, assuming that there are a total of three switching times of 1, 2, and 5, they will be destroyed and become 1066, 1066, 1066, and 1067.

[0103] S10312, reordering the processes in the first solution based on the target repair operator to obtain a second solution;

[0104] In this embodiment, the target repair operator is at least one operator selected from multiple preset repair operators in order of weight size. If the weight sizes are the same, the target repair operator is at least one operator arbitrarily selected from multiple preset repair operators. The repair operator design is designed according to the characteristics of the problem. The repair operator includes a random repair operator, that is, randomly inserting the removed process, a minimum time operator, that is, inserting in a random order at the process with the smallest switching time and waiting time, a space-time operator, that is, randomly inserting into the process with a longer waiting time in the existing process sorting, an empirical repair operator, that is, randomly extracting the historical optimal solution and inserting it according to its process order, and a constrained search operator, that is, allocating each process to an idle machine first, and performing a depth-first constrained search sort on the processes in each machine. The repair operator is a preset operator, and there are multiple repair operators. At the beginning, the weight of each repair operator is also the same.

[0105] Specifically, the number of destruction operators and repair operators is designed according to the actual situation. The repair operator and the destruction operator appear in pairs, first the destruction scheme and then the repair scheme. At the beginning, all operators have the same weight and score. The score is given in a step-by-step manner according to the different performance of the operator in each iteration process. The higher the score, the better the operator performance. The following 4 bonus situations can be set: 1 >θ 2 >θ 3 >θ 4 ≥0:

[0106] (1) A new global optimal solution is obtained after destruction / repair, +1.5 points;

[0107] (2) After destruction / repair, the global optimal solution is not obtained: ① A solution that has not been accepted but is better than the current solution is scored +1.2 points; ② A solution that has not been accepted but is worse than the current solution is scored: a) Accepting an inferior solution under certain criteria is scored +0.8 points; b) An inferior solution that does not meet the acceptance criteria is scored +0.6 points;

[0108] The step scores are set independently, and 1.5 points, 1.2 points, etc. are just examples.

[0109] During the search process, if only the best solution is accepted, it is easy to fall into the local optimum. In order to avoid this situation, certain criteria should be adopted to accept inferior solutions. In adaptive large-domain search, simulated annealing algorithm and Metropolis criterion are usually used to accept inferior solutions under a certain probability. After obtaining the new weight, the algorithm selects the operator based on the idea of ​​roulette, that is, the greater the probability of the operator, the greater the probability of being selected. Based on the above optimization algorithm and ideas, the arrangement order of the processes is iteratively improved. When the preset maximum number of iterations or the improvement effect score corresponding to the improvement scheme obtained within several consecutive iterations does not change significantly, that is, the optimal solution converges, the iteration can be stopped, and the historical optimal solution is output, that is, the scheduling solution corresponding to the maximum effect score among all the solutions of the improvement scheme obtained by historical iteration and the initial scheduling scheme is used as the historical optimal solution. Improve the production machine corresponding to each process for the historical optimal solution.

[0110] S10313, based on preset rules, change the production machine corresponding to each process in the second solution to minimize the processing time of each process and obtain an improved solution.

[0111] The improvement of the production machines corresponding to each process is processed using preset rules, and the preset rules are as follows:

[0112] Calculate the earliest possible completion time of each process i on each machine m:

[0113] t i,m = max{Completion time of the current process of equipment m t c , the completion time of the predecessor process of process i is t f}+ and the current process switching time t q + Processing time t of process i on equipment m p .

[0114] If process i is the first process of the operation, then t f = 0, if process i is the first to be processed on equipment m, then t q =0,t p = 0. Note that each process can only be processed on the specified equipment. If process i cannot be processed on equipment m, then t p = a large positive number M. The earliest possible completion time of each process on each device can be obtained. According to the known process sorting results, each process is sequentially assigned to the machine with the smallest earliest possible completion time.

[0115] For example: the production machine, i.e., equipment 001, can process the first process 1066-1 of the scheduled job 1066, with a processing speed of 3 minutes per piece, the second process 1066-2 of the scheduled job 1066, with a processing speed of 3 minutes per piece, and the first process 1067-1 of the scheduled job 1067, with a processing speed of 1 minute per piece. It takes 1 minute to switch from 1066 to 1067, and 2 minutes to switch from 1067 to 1066. Equipment 002 can process the scheduled job 1066-1. The second process 1066-2 of job 1066 has a processing speed of 5 minutes per piece, the third process 1066-3 of job 1066 has a processing speed of 4 minutes per piece, the first process 1067-1 of job 1067 has a processing speed of 2 minutes per piece, the second process 1067-2 of job 1067 has a processing speed of 2 minutes per piece, it takes 2 minutes to switch from 1066 to 1067 and 3 minutes to switch from 1067 to 1066.

[0116] If it is necessary to process the scheduled job 1066, schedule 3 processes, a total of 5 pieces, 1067, schedule 2 processes, a total of 10 pieces, code 1066, 1067, 1066, 1066, 1067, the decoding calculation time needs to select the device, the device selection will be 001, that is, 1066-1 can only be equipment 001, 002. Although 001 and 002 are both available, but at this time the device 2 is empty, the time to put 002 will be earlier, so 002 is selected, if the previous process does not affect much, then select the 001 device with short processing time per piece. After the above-mentioned optimization algorithm improvement process, the target scheduling plan is finally obtained. After the step of iterating the initial scheduling plan, it also includes, S10314, updating the weights of the target destruction operator and the target repair operator corresponding to the current improvement effect score.

[0117] In this embodiment, the destruction and repair operator is updated: According to the above description, the initial weights of each destruction operator and each repair operator are the same. A roulette strategy is adopted. In each iteration, the destruction operator and the repair operator are selected according to the weights. If the weights are the same at the beginning, they are randomly selected. After each iteration, the weight of the selected operator is updated according to the improvement effect score of the new improvement scheme in the current round. Design ω i+1 =αω i +(1-α)γ i Update the weights, where ω i+1 and ω i is the weight of the operator in the i+1th and ith rounds, α is a parameter set between 0 and 1, and γ i The design parameter θ is set according to the production scheduling score effect of this round. 1 >θ 2 >θ 3 >θ 4≥0, when the production scheduling score of this round is better than the current optimal solution, then γ i =θ 1 , when it is not better than the current optimal solution but better than the initial solution of this round, then γ i =θ 2 , when it is not better than the initial solution of this round but is accepted as the initial solution of the next round with a certain probability, then γ i =θ 3 , when it is not accepted as the initial solution for the next round, then γ i =θ 4 , ensuring that the operator with the better new feasible solution has a higher weight and has a higher probability of being selected in subsequent iterations.

[0118] By selecting destruction operators and repair operators, optimizing the arrangement order of processes, and reallocating the production machines corresponding to each process based on preset rules, the process processing time can be reduced, thereby improving the overall production efficiency. The optimization algorithm can automatically select good operators to destroy and repair the solution, with a high probability of finding a better solution. At the same time, it can avoid the occurrence of local optimal solutions as much as possible. The production machines corresponding to the process are changed based on preset rules, making the scheduling plan more flexible and adaptable to changes in production needs.

[0119] In this scheme, by selecting the destruction operator and the repair operator, the algorithm can adaptively select the most effective operator for operation, thereby enhancing the flexibility and adaptability of the algorithm; by updating the weights of the target destruction operator and the target repair operator during the iteration process, the algorithm can dynamically adjust the frequency and intensity of operator use according to the current optimization effect, thereby further improving the optimization efficiency; by iteratively updating the operator weights and selecting operators with larger weights, the algorithm can better escape from the local optimal solution and increase the probability of finding the global optimal solution.

[0120] In this embodiment, the optimization algorithm can receive the initial scheduling plan generated by the encoder, and combine the scoring model in the decoder to perform scheduling scoring, and iteratively search to obtain the optimal plan. The specific iterative process is as follows: based on the initial scheduling effect score and the improved effect score, determine whether to iterate the initial scheduling plan, including:

[0121] S10331, in response to the current improvement effect score being greater than the initial production scheduling effect score, iterating the initial scheduling solution;

[0122] S10332, in response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and the acceptance probability being greater than or equal to the preset probability, the improvement plan is replaced with the initial scheduling plan;

[0123] S10333, in response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and based on the acceptance probability being less than the preset probability, the initial scheduling plan is retained.

[0124] In this embodiment, regarding the design of the iteration steps, the optimal solution update rule and the iteration termination condition are designed with reference to the simulated annealing algorithm and the Metropolis criterion. For example, the initial temperature T set in the simulated annealing algorithm can be designed. 0 and the termination temperature T f The same initial preset parameter D0 and terminal preset parameter D1, the current temperature T increases from T 0 Starts to decay continuously, that is, the parameter D related to iteration in this application decays continuously from D0 with the number of iterations.

[0125] Design parameters Among them, F′ and F are the improvement effect score of the new improved solution obtained in the current iteration step and the initial scheduling effect score corresponding to the initial scheduling solution. When F′>F, the new improved solution searched in the current iteration step is better than the current initial solution. Here, the current initial solution refers to the initial scheduling solution in the first iteration, but after n iterations, it refers to the solution obtained at the n-1th time. Then the new improved solution is used as the initial solution for the next iteration step.

[0126] When F′≤F, the new improved solution obtained by the current iteration step is not better than the current initial solution, then with a certain probability Accept the new improved solution as the initial solution for the next iteration step with probability The acceptance probability is greater than or equal to the preset probability, which is convenient for jumping out of the local optimum. The maximum effect score obtained by all iterative steps, the initial scheduling effect score and the maximum effect score among the improved effect scores corresponding to the improved schemes obtained by all iterations and the scheduling scheme corresponding to the maximum effect score are recorded as the current optimal scheme.

[0127] If F′≤F, the new improved plan obtained by the current iteration step is not better than the current initial plan, and the acceptance probability is less than the preset probability, then the initial scheduling plan is retained.

[0128] The termination condition can be designed to reach the maximum number of iterations or the parameter D related to the iteration is less than the preset termination parameter D1 or the improvement effect score corresponding to the improved solution obtained in several consecutive iterations does not change significantly, that is, the optimal solution converges. The current optimal solution when the iteration is terminated is the optimal solution of the overall optimization algorithm.

[0129] As mentioned above, “When F′≤F, the new improved solution obtained by the current iteration step is not better than the current initial solution, then Accept the new improved solution as the initial solution for the next iteration step, so as to jump out of the local optimum. "Although the new improved solution obtained by the current iteration step search is not better than the current initial solution, that is, the improvement effect score of the new improved solution is less than the initial scheduling effect score of the initial scheduling solution, the solution may have better effects than the initial scheduling solution in some aspects. Therefore, this solution may be used as the initial solution for the next iteration step. In this solution, by iteratively comparing the initial scheduling effect score and the improvement effect score, the scheduling solution can be continuously optimized until the best or satisfactory solution is found; the solution can dynamically adjust whether to iterate according to the comparison result between the current improvement effect score and the initial scheduling effect score, thereby increasing the adaptability of the scheduling solution; through the acceptance probability mechanism in the simulated annealing algorithm, even if the current improvement effect score is not better than the initial scheduling effect score, there is a certain probability of accepting a worse solution, thereby avoiding the algorithm from converging to the local optimal solution too early; the simulated annealing algorithm balances the algorithm's exploration, finding new and possibly better solutions and using the currently found solutions by adjusting the temperature and acceptance probability, thereby increasing the possibility of finding the global optimal solution.

[0130] Specifically, the scoring model includes at least one preset scoring condition, which includes a delivery rate scoring condition, a balanced production scoring condition, and a production changeover time scoring condition. The steps of scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain the improvement effect score and the initial scheduling effect score include:

[0131] Score the improvement plan according to each preset scoring condition to obtain the improvement scoring result corresponding to each preset scoring condition; accumulate each improvement scoring result to obtain the improvement effect score. Score the initial scheduling plan according to each preset scoring condition to obtain the initial scoring result corresponding to each preset scoring condition; accumulate each initial scoring result to obtain the initial scheduling effect score.

[0132] By scoring the improvement plan and the initial scheduling plan based on at least one preset scoring condition, all aspects of the plan can be comprehensively evaluated to ensure the comprehensiveness and meticulousness of the plan optimization; the improvement scoring results are accumulated to obtain the improvement effect score, and the initial scoring results are accumulated to obtain the initial scheduling effect score, so that the improvement effect can be quantified for easy comparison and analysis; by optimizing the scheduling plan to improve the delivery rate, customer orders can be ensured to be completed on time, thereby improving customer satisfaction and loyalty; balanced production can ensure that there will be no bottlenecks in each link of the production line and improve the overall resource utilization; optimizing the conversion time can improve the production line's ability to adapt to changes in demand for different products and increase the flexibility of the production line.

[0133] Figure 3 This is a flow chart of an optimization algorithm provided in Example 1 of the present disclosure, such as Figure 3 As shown, the following will be combined Figure 3 Sorting out the processing flow of the entire optimization algorithm:

[0134] First, the initial solution generated by the encoder, that is, the initial scheduling plan, needs to be input into the optimization algorithm. If the preset termination condition is not met, iteration is performed. During the iteration, the process sorting problem is first performed, and then the production machine corresponding to each process is improved, and the improvement plan is output. The scoring model in the decoder scores the improvement plan and the initial scheduling plan to obtain the improvement effect score and the initial production scheduling effect score. According to the initial production scheduling effect score and the improvement effect score, it is determined whether to continue the iteration. The iterative process will not stop until the preset termination condition is met. Finally, the initial scheduling plan and the plan with the highest score among the improved plans are output as the target scheduling plan.

[0135] Example 2

[0136] Corresponding to the aforementioned intelligent scheduling method embodiment, the present disclosure also provides an intelligent scheduling system, which is implemented using the intelligent scheduling method as described above. Figure 4 This is a module diagram of an intelligent scheduling system provided in Embodiment 2 of the present disclosure, such as Figure 4 As shown, the intelligent scheduling system includes:

[0137] The associated information acquisition module 1 is used to acquire production associated information of multiple production machines and job associated information of multiple jobs to be scheduled; the job associated information includes all processes corresponding to each job to be scheduled;

[0138] An initial scheduling scheme generating module 2 is used to randomly generate an initial scheduling scheme by coding based on production related information and job related information;

[0139] The target scheduling solution determination module 3 is used to iteratively improve the initial scheduling solution based on the scoring model and optimization algorithm to obtain the target scheduling solution.

[0140] Preferably, the target scheduling solution determination module includes an improved solution acquisition unit, a score acquisition unit, an iteration determination unit and a target scheduling solution determination unit;

[0141] An improved plan acquisition unit is used to improve the arrangement sequence of processes in the initial scheduling plan and the production machines corresponding to each process based on an optimization algorithm to obtain an improved plan;

[0142] A scoring acquisition unit is used to score the improvement plan and the initial scheduling plan based on the scoring model to obtain the improvement effect score and the initial scheduling effect score;

[0143] An iteration determination unit, used to determine whether to iterate the initial scheduling plan based on the initial scheduling effect score and the improvement effect score;

[0144] The target scheduling scheme determining unit is used to repeat the above steps until a preset number of times is reached and / or a preset condition is satisfied, and the final initial scheduling scheme is obtained as the target scheduling scheme.

[0145] Preferably, the improved solution acquisition unit includes a first solution acquisition subunit, a second solution acquisition subunit, and an improved solution acquisition subunit;

[0146] A first plan acquisition subunit is used to destroy the arrangement order of the processes in the initial scheduling plan based on the target destruction operator to obtain a first plan;

[0147] A second solution acquisition subunit is used to reorder the processes in the first solution based on the target repair operator to obtain a second solution;

[0148] The improved solution acquisition subunit is used to change the production machine corresponding to each process in the second solution based on preset rules to minimize the processing time of each process and obtain an improved solution.

[0149] Preferably, the iterative determination unit is specifically used to iterate the initial scheduling solution in response to the current improvement effect score being greater than the initial production scheduling effect score;

[0150] The iterative determination unit is further specifically used for replacing the improvement plan with the initial scheduling plan in response to the current improvement effect score being less than or equal to the initial scheduling effect score, and obtaining the acceptance probability based on the simulated annealing algorithm, and the acceptance probability being greater than or equal to the preset probability;

[0151] The iterative determination unit is also specifically used to respond to the current improvement effect score being less than or equal to the initial scheduling effect score, and obtaining the acceptance probability based on the simulated annealing algorithm, and retaining the initial scheduling plan based on the acceptance probability being less than the preset probability.

[0152] Preferably, the target destruction operator is at least one operator selected from a plurality of preset destruction operators in order of weight size. If the weight sizes are the same, the target destruction operator is at least one operator arbitrarily selected from the plurality of preset destruction operators. The target repair operator is at least one operator selected from a plurality of preset repair operators in order of weight size. If the weight sizes are the same, the target repair operator is at least one operator arbitrarily selected from the plurality of preset repair operators.

[0153] The target scheduling scheme determination module also includes an operator weight updating unit;

[0154] The operator weight updating unit is used to update the weights of the target destruction operator and the target repair operator corresponding to the current improvement effect score.

[0155] Preferably, the scoring model includes at least one preset scoring condition, and the scoring acquisition unit includes an improved scoring result acquisition subunit, an improved effect score acquisition subunit, an initial scoring result acquisition subunit, and an initial production scheduling effect score acquisition subunit;

[0156] The improvement scoring result acquisition subunit is used to score the improvement plan according to each preset scoring condition and obtain the improvement scoring result corresponding to each preset scoring condition;

[0157] An improvement effect score obtaining subunit is used to accumulate each improvement scoring result to obtain an improvement effect score;

[0158] The initial scoring result acquisition subunit is used to score the initial scheduling plan according to each preset scoring condition to obtain the initial scoring result corresponding to each preset scoring condition;

[0159] The initial production scheduling effect score obtaining subunit is used to accumulate each initial scoring result to obtain the initial production scheduling effect score.

[0160] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, in which the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution.

[0161] Example 3

[0162] Figure 5 This is a structural diagram of an electronic device provided in Example 3 of the present disclosure, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, the intelligent scheduling method described in any of the above embodiments is implemented. Figure 5 The electronic device 50 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0163] like Figure 5As shown, the electronic device 50 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 50 may include, but are not limited to: at least one processor 51, at least one memory 52, and a bus 53 connecting different system components (including the memory 52 and the processor 51).

[0164] The bus 53 includes a data bus, an address bus, and a control bus.

[0165] The memory 52 may include a volatile memory, such as a random access memory (RAM) 521 and / or a cache memory 522 , and may further include a read-only memory (ROM) 523 .

[0166] The memory 52 may also include a program tool 525 (or utility) having a set (at least one) of program modules 524, such program modules 524 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0167] The processor 51 executes various functional applications and data processing by running the computer program stored in the memory 52, such as the intelligent scheduling method provided in any of the above embodiments.

[0168] The electronic device 50 may also communicate with one or more external devices 54 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 55. Furthermore, the electronic device 50 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 56. As shown, the network adapter 56 communicates with other modules of the electronic device 50 via a bus 53. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 50, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0169] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0170] Example 4

[0171] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the intelligent scheduling method provided by any of the above embodiments is implemented.

[0172] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0173] Example 5

[0174] The embodiment of the present disclosure also provides a computer program product, including a computer program, which implements any of the above-mentioned intelligent scheduling methods when executed by a processor.

[0175] Among them, the program code for executing the computer program product of the present disclosure can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0176] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. An intelligent scheduling method, characterized in that: The intelligent scheduling method comprises: Acquire production-related information of a plurality of production machines and job-related information of a plurality of jobs to be scheduled; the job-related information includes all processes corresponding to each of the jobs to be scheduled; Based on the production-related information and the operation-related information, randomly generating an initial scheduling plan by coding; The initial scheduling solution is iteratively improved based on the scoring model and the optimization algorithm to obtain a target scheduling solution.

2. The intelligent scheduling method according to claim 1, characterized in that: The step of iteratively improving the initial scheduling solution based on the scoring model and the optimization algorithm to obtain the target scheduling solution includes: Based on the optimization algorithm, the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process are improved to obtain an improved plan; Scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain an improvement effect score and an initial scheduling effect score; Determining whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score; Repeat the above steps until the number of iterations reaches a preset number and / or satisfies a preset condition, and the final initial scheduling scheme is obtained as the target scheduling scheme.

3. The intelligent scheduling method according to claim 2, characterized in that: The step of improving the arrangement order of the processes in the initial scheduling plan and the production machines corresponding to each process based on the optimization algorithm to obtain an improved plan includes: Destroying the arrangement order of the processes in the initial scheduling scheme based on the target destruction operator to obtain a first scheme; Reorder the processes in the first solution based on the target repair operator to obtain a second solution; Based on preset rules, the production machine corresponding to each process in the second solution is changed to shorten the processing time of each process, thereby obtaining the improved solution.

4. The intelligent scheduling method according to claim 2, characterized in that: The determining whether to iterate the initial scheduling solution based on the initial scheduling effect score and the improvement effect score includes: In response to the current improvement effect score being greater than the initial production scheduling effect score, iterating the initial scheduling solution; In response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and the acceptance probability being greater than or equal to a preset probability, the improvement plan is replaced with the initial scheduling plan; In response to the current improvement effect score being less than or equal to the initial scheduling effect score, and the acceptance probability being obtained based on the simulated annealing algorithm, and based on the acceptance probability being less than the preset probability, the initial scheduling plan is retained.

5. The intelligent scheduling method according to claim 3, characterized in that: The target destruction operator is at least one operator selected from a plurality of preset destruction operators in order of weight. If the weights are the same, the target destruction operator is at least one operator arbitrarily selected from the plurality of preset destruction operators. The target repair operator is at least one operator selected from a plurality of preset repair operators in order of weight. If the weights are the same, the target repair operator is at least one operator arbitrarily selected from the plurality of preset repair operators. After the step of iterating the initial scheduling scheme, the method further includes updating the weights of the target destruction operator and the target repair operator corresponding to the current improvement effect score.

6. The intelligent scheduling method according to claim 2, characterized in that: The scoring model includes at least one preset scoring condition, and the step of scoring the improvement plan and the initial scheduling plan based on the scoring model to obtain an improvement effect score and an initial scheduling effect score includes: Scoring the improvement plan according to each of the preset scoring conditions to obtain an improvement scoring result corresponding to each of the preset scoring conditions; Accumulating each of the improvement scoring results to obtain the improvement effect score; Scoring the initial scheduling scheme according to each of the preset scoring conditions to obtain an initial scoring result corresponding to each of the preset scoring conditions; Each of the initial scoring results is accumulated to obtain the initial production scheduling effect score.

7. The intelligent scheduling method according to claim 6, characterized in that: The preset scoring conditions include a delivery rate scoring condition, a balanced production scoring condition and a production changeover time scoring condition.

8. An intelligent scheduling system, characterized in that: The intelligent scheduling system comprises: A correlation information acquisition module, used to acquire production correlation information of multiple production machines and job correlation information of multiple jobs to be scheduled; the job correlation information includes all processes corresponding to each of the jobs to be scheduled; An initial scheduling scheme generating module, used to randomly generate an initial scheduling scheme by coding based on the production-related information and the operation-related information; The target scheduling scheme determination module is used to iteratively improve the initial scheduling scheme based on the scoring model and the optimization algorithm to obtain the target scheduling scheme.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, the intelligent scheduling method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the intelligent scheduling method according to any one of claims 1 to 7 is implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the intelligent scheduling method according to any one of claims 1 to 7 is implemented.

Images