A method and device for constructing a production scheduling model, a method and device for production scheduling, an equipment and a medium

Abstract

The present disclosure provides a production scheduling model construction method, a production scheduling method, a device, equipment and a medium, relates to the field of intelligent industry and data processing technology, in particular to the field of workshop production scheduling technology; the method comprises the following steps: adding at least one dimension of a preset attribute parameter to a variable in a solver; collecting a plurality of groups of sample production scheduling data and sample production scheduling results corresponding to each production scheduling problem in a flexible job shop; the sample production scheduling data comprises a plurality of preset solving variables, the attribute parameters of the preset solving variables correspond to a basic attribute parameter and at least one dimension of a preset attribute parameter; based on the sample production scheduling data and the corresponding sample production scheduling results, a target production scheduling model for outputting a production scheduling result representing a production scheduling optimization scheme is constructed. The present disclosure is particularly suitable for the solver acceleration solving scene of large-scale flexible job shop production scheduling problems, can effectively reduce the solving complexity of the solver, improve the solving speed, and ensure that the workshop production scheduling is reasonable and orderly executed.

Description

Technical Field

[0001] This disclosure relates to the fields of smart industry and data processing technology, and in particular to the field of workshop scheduling technology, providing a scheduling model construction method, scheduling method, device, equipment and medium. Background Technology

[0002] Flexible shop floor scheduling (such as automobile production scheduling, machine tool processing production scheduling, etc.) is an NP-hard problem (a type of difficult problem). Each order in the shop floor has a series of processes, which must be executed in a certain order. Each process requires selecting a piece of equipment that meets the requirements of that process, and also requires selecting a start time for that process.

[0003] Currently, flexible workshop scheduling problems are mainly solved using purely heuristic methods (such as population evolution combined with neighborhood search) or methods based on pure solvers. However, all of these solutions have the problem that the scheduling of order processes cannot be reasonable and orderly. For example, for large-scale flexible workshop scheduling problems, it is difficult to solve them using pure solvers, and no feasible solution can be found, which can easily lead to order delivery delays, i.e., failure to meet the order deadline. Summary of the Invention

[0004] The technical problem to be solved by this disclosure is to overcome the defects in the prior art and provide a production scheduling model construction method, production scheduling method, apparatus, equipment and medium.

[0005] This disclosure solves the above-mentioned technical problems through the following technical solution:

[0006] One aspect of this disclosure provides a method for constructing a production scheduling model, the method comprising:

[0007] Add at least one preset attribute parameter to the variables in the solver;

[0008] The preset attribute parameters are parameters that assist the solver in obtaining feasible solutions;

[0009] Collect several sets of sample production scheduling data corresponding to each type of production scheduling problem in the flexible operation workshop, and obtain the sample production scheduling results corresponding to the sample production scheduling data;

[0010] The sample production scheduling data includes multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to basic attribute parameters and at least one dimension of the preset attribute parameters.

[0011] Based on the sample production scheduling data and the corresponding sample production scheduling results, a target production scheduling model is constructed to output the production scheduling results that characterize the production scheduling optimization scheme.

[0012] In another aspect, this disclosure also provides a production scheduling method for a flexible workshop, the production scheduling method being implemented based on a target production scheduling model obtained by the above-described production scheduling model construction method, the production scheduling method comprising:

[0013] Collect several sets of actual production scheduling data corresponding to different scheduling problems in each production scheduling task in the flexible operation workshop;

[0014] Each set of actual production scheduling data is input into the target production scheduling model to obtain the output predicted production scheduling result;

[0015] The target production scheduling model is solved using a solver to obtain the solution results;

[0016] Based on the predicted production schedule results and the solution results, the corresponding preset solution variables included in each group of actual production schedule data are determined to be fixed variables;

[0017] Obtain the other solution variables besides the fixed variables from all the preset solution variables corresponding to each group of actual production scheduling data;

[0018] The solver is used to solve the other variables to obtain the target production schedule result for each set of actual production schedule data.

[0019] In another aspect of this disclosure, a production scheduling model building apparatus is also provided, the apparatus comprising:

[0020] The attribute parameter addition module is used to add at least one preset attribute parameter to the variables in the solver.

[0021] The preset attribute parameters are parameters that assist the solver in obtaining feasible solutions;

[0022] The sample scheduling data acquisition module is used to collect several sets of sample scheduling data corresponding to each scheduling problem in the flexible operation workshop.

[0023] The sample production scheduling result acquisition module is used to acquire the sample production scheduling result corresponding to the sample production scheduling data;

[0024] The sample production scheduling data includes multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to basic attribute parameters and at least one dimension of the preset attribute parameters.

[0025] The target production scheduling model construction module is used to construct a target production scheduling model based on the sample production scheduling data and the corresponding sample production scheduling results, which is used to output the production scheduling results that characterize the production scheduling optimization scheme.

[0026] In another aspect, this disclosure also provides a production scheduling device for a flexible workshop, wherein the production scheduling method is implemented based on a target production scheduling model obtained by the aforementioned production scheduling model construction device, and the production scheduling device includes:

[0027] The actual production scheduling data acquisition module is used to collect several sets of actual production scheduling data corresponding to different scheduling problems in each scheduling task in the flexible operation workshop.

[0028] The production scheduling result prediction module is used to input each set of actual production scheduling data into the target production scheduling model to obtain the output predicted production scheduling result;

[0029] The solution result acquisition module is used to solve the target production scheduling model using a solver to obtain the solution result;

[0030] The fixed variable determination module is used to determine the corresponding preset solution variables included in each group of actual production data as fixed variables based on the predicted production scheduling results and the solution results.

[0031] Other solution variable acquisition module is used to acquire other solution variables besides the fixed variables among all the preset solution variables corresponding to each group of actual production scheduling data;

[0032] The target production scheduling result acquisition module is used to use the solver to solve the other solution variables to obtain the target production scheduling result for each group of actual production scheduling data.

[0033] In another aspect of this disclosure, an electronic device is also provided, comprising:

[0034] At least one processor; and

[0035] A memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method described above.

[0037] In another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform the method according to the above description.

[0038] In another aspect of this disclosure, a computer program product is also provided, comprising a computer program that, when executed by a processor, implements the method described above.

[0039] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0040] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0041] Figure 1 This is a flowchart of the production scheduling model construction method according to the first embodiment of this disclosure.

[0042] Figure 2 This is a flowchart of the production scheduling model construction method according to the first embodiment of this disclosure.

[0043] Figure 3 This is a flowchart of the production scheduling model construction method according to the first embodiment of this disclosure.

[0044] Figure 4 This is a schematic diagram illustrating the production scheduling model construction of the first embodiment of this disclosure.

[0045] Figure 5 This is a flowchart of a production scheduling method for a flexible workshop according to a second embodiment of this disclosure.

[0046] Figure 6 This is a flowchart of a production scheduling method for a flexible workshop according to a second embodiment of this disclosure.

[0047] Figure 7 This is a flowchart of a production scheduling method for a flexible workshop according to a second embodiment of this disclosure.

[0048] Figure 8 This is a schematic diagram illustrating the solution of fixed variables in the second embodiment of this disclosure.

[0049] Figure 9 This is a schematic diagram of the modules of the production scheduling model construction method according to the third embodiment of this disclosure.

[0050] Figure 10 This is a schematic diagram of the modules of the production scheduling model construction method according to the third embodiment of this disclosure.

[0051] Figure 11 This is a schematic diagram of the production scheduling method for a flexible workshop according to the fourth embodiment of this disclosure.

[0052] Figure 12 This is a schematic diagram of the production scheduling method for a flexible workshop according to the fourth embodiment of this disclosure.

[0053] Figure 13 This is a schematic diagram of the structure of an electronic device according to the fifth embodiment of this disclosure. Detailed Implementation

[0054] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0055] Example 1

[0056] like Figure 1 As shown, the production scheduling model construction method in this embodiment includes:

[0057] S101. Add at least one preset attribute parameter to the variables in the solver;

[0058] Among them, the preset attribute parameters are the parameters used by the auxiliary solver to obtain feasible solutions;

[0059] Specifically, the preset attribute parameters are attribute parameters other than the basic attribute parameters of variables in the solver. The preset attribute parameters include, but are not limited to, group variable group attributes and step priority attributes.

[0060] For example, in a flexible workshop scheduling scenario, there are two basic solution variables: X[i,j]: whether process i selects equipment j for processing; Y[i,j]: the start time of process i on equipment j. Specifically, for the solution variable X[i,j], by adding a preset attribute parameter corresponding to X[i,j], assigning it to a specified group variable group and setting it to the highest priority, the solver can ultimately obtain a feasible solution corresponding to the solution variable for how process i selects equipment in the corresponding troubleshooting task, thereby achieving the purpose of troubleshooting optimization.

[0061] S102. Collect several sets of sample production scheduling data corresponding to each type of production scheduling problem in the flexible operation workshop;

[0062] Flexible work workshops include, but are not limited to, automobile production scheduling and machine tool processing production scheduling.

[0063] S103. Obtain the sample production scheduling results corresponding to the sample production scheduling data;

[0064] The sample production scheduling data contains multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to the basic attribute parameters and at least one dimension of preset attribute parameters.

[0065] S104. Based on the sample production scheduling data and the corresponding sample production scheduling results, construct a target production scheduling model to output the production scheduling results that characterize the production scheduling optimization scheme.

[0066] In this embodiment, by modifying the modeling language, additional attribute parameters are added to each integer variable in addition to the basic attribute parameters when designing variables. The model is trained based on these variable solutions, which ensures the accuracy and reliability of the target production scheduling model used to output the production scheduling plan, thereby effectively improving the rationality and orderliness of process execution in the subsequent flexible operation workshop.

[0067] like Figure 2 As shown, the production scheduling model construction method in this embodiment is... Figure 1 Further improvements to the technical solution, specifically:

[0068] In one feasible embodiment, step S101 includes:

[0069] S1011. Based on the basic attribute parameters of variables in the solver, add corresponding variable groups and priorities.

[0070] In a feasible solution, the construction method also includes:

[0071] Get the attribute parameter adjustment instructions;

[0072] Based on the attribute parameter adjustment command, the variable in the solver can be set with other attribute parameters besides the basic attribute parameters by using at least one of the following adjustment methods: modification, addition, or deletion.

[0073] This solution allows for flexible setting of attribute parameters, better meeting the needs of more demanding scheduling optimization scenarios. Other attribute parameters can also be added according to actual requirements, as long as they meet the solution requirements. This will not be elaborated further here.

[0074] In one feasible embodiment, step S102 includes:

[0075] S1021. Use a bipartite diagram to collect several sets of sample production scheduling data corresponding to each production scheduling problem in the flexible workshop;

[0076] In the flexible work workshop, different scheduling tasks correspond to different scheduling problems. A bipartite graph is used to collect n sets of obs data (one type of scheduling data) corresponding to various types of scheduling problems. For each sub-problem under each type of scheduling problem, there is a set of obs data, which is used for subsequent model training.

[0077] Furthermore, the method for collecting data in bipartite diagrams is a mature technology in this field, and therefore will not be elaborated upon here. It should be noted that other methods capable of effectively collecting scheduling data for each scheduling problem in a flexible workshop can also be used, without specific limitations.

[0078] Step S103 includes:

[0079] S1031. Obtain the corresponding sample production schedule result based on the production schedule result identifier information corresponding to the sample production schedule data.

[0080] The production scheduling data used for model training all have matching production scheduling results. It is necessary to label these production scheduling data to establish the correlation between the production scheduling data and the corresponding production scheduling results. This process can be generated automatically or implemented manually. If an automatic method is used, the processing efficiency and accuracy can be effectively guaranteed, thereby ensuring the accuracy and reliability of subsequent model training.

[0081] In one feasible embodiment, step S104 includes:

[0082] S1041. Based on the sample production scheduling data and the corresponding sample production scheduling results, construct the first production scheduling model for the corresponding production scheduling problem;

[0083] S1042. Based on the first production scheduling model for different production scheduling problems, obtain the initial production scheduling model;

[0084] For each type of production scheduling problem, a corresponding first production scheduling model needs to be trained based on the sample production scheduling data. For example, if the different types of production scheduling problems include A, B, and C, then the corresponding production scheduling models for problems A, B, and C are constructed respectively.

[0085] The initial production scheduling model is a comprehensive or integrated model obtained by summarizing and processing the model parameters of these different types of production scheduling problems.

[0086] Furthermore, there are no special restrictions on which network model to use to train the corresponding first production model, as long as the accuracy of the first production model for the corresponding problem can meet the set training conditions.

[0087] S1043. Use a solver to solve the sample production scheduling data and obtain the corresponding first production scheduling result;

[0088] S1044. Based on the sample production scheduling data and the first production scheduling result, retrain the first production scheduling model to optimize the initial production scheduling model until the preset optimization conditions are met to obtain the final target production scheduling model.

[0089] In this scheme, based on the priority of different preset solution variables in the sample production scheduling data, the solver processes the sample production scheduling data according to priority to obtain the scheduling results as the output of model training. Steps S1041-S1044 are repeatedly executed until a model that meets the preset optimization conditions is obtained as the final target production scheduling model, thereby effectively ensuring the accuracy and reliability of model training. The preset optimization conditions can be pre-set based on actual experience.

[0090] In a feasible solution, such as Figure 3 As shown, step S1042 specifically includes:

[0091] S10421. A preset fusion method is used to fuse the first production scheduling models of different production scheduling problems to obtain an initial production scheduling model. That is, the model parameters of the first production scheduling models of different production scheduling problems are summarized and processed to finally obtain a fused comprehensive production scheduling model.

[0092] In one feasible solution, step S1043 specifically includes:

[0093] S10431. The solver is used to differentiate and process the preset solver variables in the sample production scheduling data according to their priorities, so as to obtain the first production scheduling result corresponding to the preset solver variables of different priorities in the sample production scheduling data.

[0094] Specifically, such as Figure 4 As shown, for example, different types of scheduling problems include A and B, respectively constructing the corresponding A-problem scheduling model and the B-problem scheduling model; the model parameters of the A-problem scheduling model and the B-problem scheduling model are fused to obtain the initial scheduling model and stored in the model pool;

[0095] Based on the priority of different preset solution variables in the sample production scheduling data, the solver solves the sample production scheduling data according to priority to obtain the scheduling results as the output of model training. Based on the output of the solver, the output of the corresponding sample production scheduling data for model training is re-determined. Then, the A-problem production scheduling model for problem A and the B-problem production scheduling model for problem B are retrained. The model parameters of the retrained A-problem production scheduling model and the B-problem production scheduling model are fused again to obtain the optimized initial scheduling model and store it in the model pool.

[0096] The above process of retraining the model is repeated until the model meets the set conditions. Then, the training of the model is stopped to obtain the target production scheduling model that meets the requirements.

[0097] Example 2

[0098] like Figure 5As shown, the production scheduling method of the flexible workshop in this embodiment is implemented based on the target production scheduling model obtained by the production scheduling model construction method in Embodiment 1.

[0099] The production scheduling method in this embodiment includes:

[0100] S201. Collect several sets of actual production scheduling data corresponding to different production scheduling problems in each production scheduling task in the flexible operation workshop;

[0101] Specifically, a bipartite graph was used to collect n sets of obs data corresponding to different types of scheduling problems under different scheduling tasks in the flexible work workshop; for each type of scheduling problem, each sub-problem corresponds to a set of obs data, which is used for subsequent model training.

[0102] Of course, other methods that can effectively collect scheduling data for each scheduling problem in a flexible workshop can also be used, without specific limitations.

[0103] S202. Input each set of actual production scheduling data into the target production scheduling model to obtain the output predicted production scheduling results;

[0104] S203. Use a solver to solve the target production scheduling model and obtain the solution results;

[0105] S204. Based on the predicted production schedule results and the solution results, determine the corresponding preset solution variables included in each group of actual production schedule data as fixed variables;

[0106] In this process, the solver uses a heuristic to fix variables each time, and each fixation uses grouping. Only one group of variables is fixed each time until all the fixed variables are obtained, in order to ensure faster solving in subsequent steps.

[0107] S205. Obtain the other solution variables besides the fixed variables from all preset solution variables corresponding to each set of actual production scheduling data;

[0108] S206. Use a solver to solve other variables to obtain the target production schedule result for each set of actual production scheduling data.

[0109] In this embodiment, for several sets of actual production scheduling data corresponding to different production scheduling problems, the preset solution variables in each set of actual production scheduling data are given additional variable group attributes and priority attributes. Fixed variables are obtained by solving the target production scheduling model of the solver. These fixed variables do not participate in the subsequent solution process. Only the remaining solution variables participate in the solution and ensure that a feasible solution can be found. This reduces the solution complexity of the solver and greatly improves the solution speed. It is particularly suitable for the accelerated solution scenario of large-scale flexible workshop production scheduling problems. It ensures the reasonable and orderly execution of processes in flexible workshop production scheduling and can effectively reduce the delay value to better meet the deadline requirements of production orders. This achieves the purpose of reducing order delivery delays and increasing work-in-process turnover, overcoming the defects of low efficiency and easy error in manual production scheduling.

[0110] The flexible workshop scheduling method described above can be well applied when solvers such as gurobi (mathematical programming optimizer) cannot find a feasible solution for large-scale scheduling problems, ensuring that a feasible solution can be found as soon as possible, thus achieving the effect of scheduling and optimization.

[0111] The production scheduling method of the flexible workshop in this embodiment is Figure 3 Further improvements to the technical solution, specifically:

[0112] In a feasible solution, such as Figure 6 As shown, step S204 includes:

[0113] S20411. When the solution results indicate the existence of a feasible solution, the predicted production scheduling results and the solution results are hybridized to obtain the first hybridization result.

[0114] S20412. Based on the results of the first hybridization, the corresponding preset solution variables included in the actual production data of each group are determined to be fixed variables.

[0115] In a feasible solution, such as Figure 7 As shown, step S204 includes:

[0116] S20421. If no feasible solution exists in the solution results, then obtain the linear programming LP solution from the solver.

[0117] S20422. Perform hybridization processing on the predicted production scheduling results and the linear programming LP solution to obtain the second hybridization result;

[0118] S20423. Based on the results of the second hybridization, the corresponding preset solution variables included in the actual production data of each group are determined to be fixed variables.

[0119] Specifically, such as Figure 8As shown, in the actual production scheduling scenario of a flexible workshop, if the solver's solution indicates the existence of a current optimal solution / other feasible solutions, then the predicted production scheduling result output by the model is hybridized with the solution result to obtain the corresponding fixed variables. If the solver's solution indicates the absence of a current optimal solution / other feasible solutions, then the predicted production scheduling result output by the model is hybridized with the current linear programming (LP) solution in the solver to obtain the corresponding fixed variables. This production scheduling scheme ensures that a feasible solution can be obtained regardless of the situation, i.e., by modifying the hybridization logic of the RINS enhancement heuristic and adding the RINS calling logic of the solver, the purpose of effective production scheduling optimization is achieved.

[0120] Example 3

[0121] like Figure 9 As shown, the production scheduling model construction device in this embodiment includes:

[0122] Attribute parameter addition module 1 is used to add at least one preset attribute parameter of a dimension to the variables in the solver;

[0123] Among them, the preset attribute parameters are the parameters used by the auxiliary solver to obtain feasible solutions;

[0124] Specifically, the preset attribute parameters are attribute parameters other than the basic attribute parameters of variables in the solver. The preset attribute parameters include, but are not limited to, group variable group attributes and step priority attributes.

[0125] For example, in a flexible workshop scheduling scenario, there are two basic solution variables: X[i,j]: whether process i selects equipment j for processing; Y[i,j]: the start time of process i on equipment j. Specifically, for the solution variable X[i,j], by adding a preset attribute parameter corresponding to X[i,j], assigning it to a specified group variable group and setting it to the highest priority, the solver can ultimately obtain a feasible solution corresponding to the solution variable for how process i selects equipment in the corresponding troubleshooting task, thereby achieving the purpose of troubleshooting optimization.

[0126] Sample scheduling data acquisition module 2 is used to collect several sets of sample scheduling data corresponding to each scheduling problem in the flexible operation workshop;

[0127] Flexible work workshops include, but are not limited to, automobile production scheduling and machine tool processing production scheduling.

[0128] Sample scheduling result acquisition module 3 is used to acquire the sample scheduling result corresponding to the sample scheduling data;

[0129] The sample production scheduling data contains multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to the basic attribute parameters and at least one dimension of preset attribute parameters.

[0130] The target production scheduling model construction module 4 is used to construct a target production scheduling model based on sample production scheduling data and corresponding sample production scheduling results, which is used to output the production scheduling results that characterize the production scheduling optimization scheme.

[0131] In this embodiment, by modifying the modeling language, additional attribute parameters are added to each integer variable in addition to the basic attribute parameters when designing variables. The model is trained based on these variable solutions, which ensures the accuracy and reliability of the target production scheduling model used to output the production scheduling plan, thereby effectively improving the rationality and orderliness of process execution in the subsequent flexible operation workshop.

[0132] like Figure 10 As shown, the production scheduling model construction device in this embodiment is Figure 9 Further improvements to the technical solution, specifically:

[0133] In one feasible solution, the attribute parameter addition module 1 is used to add corresponding variable groups and priorities based on the basic attribute parameters of variables in the solver.

[0134] In one feasible embodiment, the construction apparatus also includes:

[0135] Parameter adjustment instruction acquisition module 5 is used to acquire attribute parameter adjustment instructions;

[0136] The attribute parameter adjustment module 6 is used to set other attribute parameters of variables in the solver besides the basic attribute parameters, based on the attribute parameter adjustment command, using at least one of the adjustment methods of modification, addition, and deletion.

[0137] This solution allows for flexible setting of attribute parameters, better meeting the needs of more demanding scheduling optimization scenarios. Other attribute parameters can also be added according to actual requirements, as long as they meet the solution requirements. This will not be elaborated further here.

[0138] In one feasible solution, the sample scheduling data acquisition module 2 is used to acquire several sets of sample scheduling data corresponding to each scheduling problem in the flexible workshop using a bipartite diagram;

[0139] In the flexible operation workshop, different scheduling tasks correspond to different scheduling problems. A bipartite graph is used to collect n sets of obs data corresponding to various types of scheduling problems. For each type of scheduling problem, each sub-problem corresponds to a set of obs data, which is used for subsequent model training.

[0140] Furthermore, the method for collecting data in bipartite diagrams is a mature technology in this field, and therefore will not be elaborated upon here. It should be noted that other methods capable of effectively collecting scheduling data for each scheduling problem in a flexible workshop can also be used, without specific limitations.

[0141] The sample production scheduling result acquisition module 3 is used to obtain the corresponding sample production scheduling result based on the production scheduling result identification information corresponding to the sample production scheduling data.

[0142] The production scheduling data used for model training all have matching production scheduling results. It is necessary to label these production scheduling data to establish the correlation between the production scheduling data and the corresponding production scheduling results. This process can be generated automatically or implemented manually. If an automatic method is used, the processing efficiency and accuracy can be effectively guaranteed, thereby ensuring the accuracy and reliability of subsequent model training.

[0143] In a feasible solution, the target scheduling model construction module 4 includes:

[0144] The first production scheduling model construction unit 7 is used to construct the first production scheduling model for the corresponding production scheduling problem based on the sample production scheduling data and the corresponding sample production scheduling results.

[0145] Initial production scheduling model construction unit 8 is used to obtain the initial production scheduling model based on the first production scheduling model for different production scheduling problems;

[0146] The first production scheduling result acquisition unit 9 is used to solve the sample production scheduling data using a solver to obtain the corresponding first production scheduling result;

[0147] The target production scheduling module acquisition unit 10 is used to retrain the first production scheduling model based on the sample production scheduling data and the first production scheduling result to optimize the initial production scheduling model until the preset optimization conditions are met to obtain the final target production scheduling model.

[0148] For each type of production scheduling problem, a corresponding first production scheduling model needs to be trained based on the sample production scheduling data. For example, if the different types of production scheduling problems include A, B, and C, then the corresponding production scheduling models for problems A, B, and C are constructed respectively.

[0149] The initial production scheduling model is a comprehensive or integrated model obtained by summarizing and processing the model parameters of these different types of production scheduling problems.

[0150] Furthermore, there are no special restrictions on which network model to use to train the corresponding first production model, as long as the accuracy of the first production model for the corresponding problem can meet the set training conditions.

[0151] In one feasible solution, the initial production scheduling model construction unit 8 is used to merge the first production scheduling models of different production scheduling problems using a preset fusion method to obtain the initial production scheduling model.

[0152] That is, the model parameters of the first production scheduling model for different production scheduling problems are summarized and processed to finally obtain a comprehensive production scheduling model after fusion.

[0153] In one feasible scheme, the first production scheduling result acquisition unit 9 is used to use a solver to distinguish and process different preset solver variables in the sample production scheduling data according to their priorities, so as to obtain the first production scheduling result corresponding to the preset solver variables of different priorities in the sample production scheduling data respectively.

[0154] Specifically, such as Figure 4 As shown, for example, different types of scheduling problems include A and B, respectively constructing the corresponding A-problem scheduling model and the B-problem scheduling model; the model parameters of the A-problem scheduling model and the B-problem scheduling model are fused to obtain the initial scheduling model and stored in the model pool;

[0155] Based on the priority of different preset solution variables in the sample production scheduling data, the solver solves the sample production scheduling data according to priority to obtain the scheduling results as the output of model training. Based on the output of the solver, the output of the corresponding sample production scheduling data for model training is re-determined. Then, the A-problem production scheduling model for problem A and the B-problem production scheduling model for problem B are retrained. The model parameters of the retrained A-problem production scheduling model and the B-problem production scheduling model are fused again to obtain the optimized initial scheduling model and store it in the model pool.

[0156] The above process of retraining the model is repeated until the model meets the set conditions. Then, the training of the model is stopped to obtain the target production scheduling model that meets the requirements.

[0157] Example 4

[0158] like Figure 11 As shown, the production scheduling device of the flexible workshop in this embodiment is based on the target production scheduling model obtained by the production scheduling model construction device in embodiment 3.

[0159] The production scheduling device in this embodiment includes:

[0160] The actual production scheduling data acquisition module 11 is used to collect several sets of actual production scheduling data corresponding to different production scheduling problems in each production scheduling task in the flexible operation workshop.

[0161] Specifically, a bipartite graph was used to collect n sets of obs data corresponding to different types of scheduling problems under different scheduling tasks in the flexible work workshop; for each type of scheduling problem, each sub-problem corresponds to a set of obs data, which is used for subsequent model training.

[0162] Of course, other methods that can effectively collect scheduling data for each scheduling problem in a flexible workshop can also be used, without specific limitations.

[0163] The production scheduling result prediction module 12 is used to input each set of actual production scheduling data into the target production scheduling model in order to obtain the output predicted production scheduling result.

[0164] The solution result acquisition module 13 is used to solve the target production scheduling model using a solver and obtain the solution result;

[0165] In this process, the solver uses a heuristic to fix variables each time, and each fixation uses grouping. Only one group of variables is fixed each time until all the fixed variables are obtained, in order to ensure faster solving in subsequent steps.

[0166] The fixed variable determination module 14 is used to determine the corresponding preset solution variables included in each group of actual production scheduling data as fixed variables based on the predicted production scheduling results and the solution results.

[0167] Other solution variable acquisition module 15 is used to acquire other solution variables besides fixed variables among all preset solution variables corresponding to each group of actual production scheduling data;

[0168] The target production scheduling result acquisition module 16 is used to use a solver to solve other solution variables in order to obtain the target production scheduling result for each set of actual production scheduling data.

[0169] In this embodiment, for several sets of actual production scheduling data corresponding to different production scheduling problems, the preset solution variables in each set of actual production scheduling data are given additional variable group attributes and priority attributes. Fixed variables are obtained by solving the target production scheduling model of the solver. These fixed variables do not participate in the subsequent solution process. Only the remaining solution variables participate in the solution and ensure that a feasible solution can be found. This reduces the solution complexity of the solver and greatly improves the solution speed. It is particularly suitable for the accelerated solution scenario of large-scale flexible workshop production scheduling problems. It ensures the reasonable and orderly execution of processes in flexible workshop production scheduling and can effectively reduce the delay value to better meet the deadline requirements of production orders. This achieves the purpose of reducing order delivery delays and increasing work-in-process turnover, overcoming the defects of low efficiency and easy error in manual production scheduling.

[0170] The flexible workshop scheduling method described above can be well applied when solvers such as gurobi cannot find a feasible solution for large-scale scheduling problems, ensuring that a feasible solution can be found as soon as possible, thus achieving the effect of troubleshooting and optimization.

[0171] like Figure 12 As shown, the production scheduling device for the flexible workshop in this embodiment is Figure 11 Further improvements to the technical solution, specifically:

[0172] In one feasible embodiment, the fixed variable determination module 14 includes:

[0173] The first hybridization result acquisition unit 17 is used to perform hybridization processing on the predicted production scheduling result and the solution result when the solution result indicates that there is a feasible solution, and to obtain the first hybridization result;

[0174] The first fixed variable determination unit 18 is used to determine the corresponding preset solution variables included in each group of actual production data as fixed variables based on the first hybridization result.

[0175] In one feasible scheme, the second hybridization result acquisition unit 19 is used to acquire the linear programming LP solution in the solver when there is no feasible solution in the solution result, and to perform hybridization processing on the predicted production scheduling result and the linear programming LP solution to obtain the second hybridization result;

[0176] The second fixed variable determination unit 20 is used to determine the corresponding preset solution variables included in each group of actual production data as fixed variables based on the second hybridization results.

[0177] Specifically, such as Figure 8 As shown, in the actual production scheduling scenario of a flexible workshop, if the solver's solution indicates the existence of a current optimal solution / other feasible solutions, then the predicted production scheduling result output by the model is hybridized with the solution result to obtain the corresponding fixed variables. If the solver's solution indicates the absence of a current optimal solution / other feasible solutions, then the predicted production scheduling result output by the model is hybridized with the current linear programming (LP) solution in the solver to obtain the corresponding fixed variables. This production scheduling scheme ensures that a feasible solution can be obtained regardless of the situation, i.e., by modifying the hybridization logic of the RINS enhancement heuristic and adding the RINS calling logic of the solver, the purpose of effective production scheduling optimization is achieved.

[0178] Example 5

[0179] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0180] Figure 13A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0181] like Figure 13 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0182] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of monitors, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0183] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the methods described above. For example, in some embodiments, the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the methods described above by any other suitable means (e.g., by means of firmware).

[0184] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0185] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0186] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0187] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0188] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0189] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0190] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0191] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for constructing a production scheduling model, the method comprising: In addition to the basic attribute parameters of the variables in the solver, at least one preset attribute parameter of one dimension is added. The preset attribute parameters are parameters that assist the solver in obtaining feasible solutions, and the preset attribute parameters in at least one dimension include variable groups and priorities. Collect several sets of sample production scheduling data corresponding to each type of production scheduling problem in the flexible operation workshop, and obtain the sample production scheduling results corresponding to the sample production scheduling data; The sample production scheduling data includes multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to the basic attribute parameters and the preset attribute parameters of at least one dimension. Based on the sample production scheduling data and the corresponding sample production scheduling results, a target production scheduling model is constructed to output the production scheduling results that characterize the production scheduling optimization scheme. The step of constructing a target production scheduling model based on the sample production scheduling data and the corresponding sample production scheduling results to output a production scheduling result characterizing the production scheduling optimization scheme includes: Based on the sample production scheduling data and the corresponding sample production scheduling results, a first production scheduling model is constructed for the corresponding production scheduling problem. Based on the first production scheduling model for different production scheduling problems, an initial production scheduling model is obtained; The solver is used to process the sample production scheduling data to obtain the corresponding first production scheduling result; Based on the sample production scheduling data and the first production scheduling result, the first production scheduling model is retrained to optimize the initial production scheduling model until the preset optimization conditions are met to obtain the final target production scheduling model.

2. The production scheduling model construction method as described in claim 1, further comprising: Get the attribute parameter adjustment instructions; Based on the attribute parameter adjustment instructions, the variables in the solver are set with other attribute parameters besides the basic attribute parameters by means of modification, addition, or deletion.

3. The production scheduling model construction method as described in claim 1, wherein the step of using the solver to solve the sample production scheduling data and obtaining the corresponding first production scheduling result includes: The solver is used to differentiate and process the preset solver variables in the sample production data according to their priorities, so as to obtain the first production schedule result corresponding to the preset solver variables with different priorities in the sample production data.

4. The production scheduling model construction method as described in claim 1, wherein the step of collecting several sets of sample production scheduling data corresponding to each production scheduling problem in the flexible workshop includes: A bipartite graph is used to collect several sets of sample production scheduling data corresponding to each of the production scheduling problems in the flexible work workshop; The step of obtaining the sample production scheduling result corresponding to the sample production scheduling data includes: The corresponding sample production schedule result is obtained based on the production schedule result identifier information corresponding to the sample production schedule data.

5. The production scheduling model construction method as described in claim 1, wherein the step of obtaining an initial production scheduling model based on the first production scheduling model for different production scheduling problems includes: The first scheduling models for different scheduling problems are fused using a preset fusion method to obtain the initial scheduling model.

6. A production scheduling method for a flexible workshop, wherein the production scheduling method is implemented based on a target production scheduling model obtained by the production scheduling model construction method according to any one of claims 1-5, the production scheduling method comprising: Collect several sets of actual production scheduling data corresponding to different scheduling problems in each production scheduling task in the flexible operation workshop; Each set of actual production scheduling data is input into the target production scheduling model to obtain the output predicted production scheduling result; The target production scheduling model is solved using a solver to obtain the solution results; Based on the predicted production schedule results and the solution results, the corresponding preset solution variables included in each group of actual production schedule data are determined to be fixed variables; Obtain the other solution variables besides the fixed variables from all the preset solution variables corresponding to each group of actual production scheduling data; The solver is used to solve the other variables to obtain the target production schedule result for each set of actual production schedule data.

7. The production scheduling method for a flexible workshop as described in claim 6, wherein the step of determining, based on the predicted production scheduling results and the solution results, that the corresponding preset solution variables included in each group of actual production scheduling data are fixed variables, comprises: When the solution result indicates that a feasible solution exists, the predicted production scheduling result and the solution result are hybridized to obtain a first hybridization result; Based on the first hybridization result, the corresponding preset solution variable included in each group of actual production scheduling data is determined to be the fixed variable.

8. The production scheduling method for a flexible workshop as described in claim 6, wherein the step of determining, based on the predicted production scheduling result and the solution result, that the corresponding preset solution variable included in each group of actual production scheduling data is a fixed variable, comprises: If no feasible solution is found in the solution results, then the linear programming LP solution in the solver is obtained; The predicted production schedule result and the linear programming LP solution are hybridized to obtain a second hybridization result; Based on the second hybridization result, the corresponding preset solution variable included in each group of actual production data is determined to be the fixed variable.

9. A production scheduling model construction device, the construction device comprising: The attribute parameter addition module is used to add at least one preset attribute parameter in a dimension to the basic attribute parameters of variables in the solver. The preset attribute parameters are parameters that assist the solver in obtaining feasible solutions, and the preset attribute parameters in at least one dimension include variable groups and priorities. The sample scheduling data acquisition module is used to collect several sets of sample scheduling data corresponding to each scheduling problem in the flexible operation workshop. The sample production scheduling result acquisition module is used to acquire the sample production scheduling result corresponding to the sample production scheduling data; The sample production scheduling data includes multiple preset solution variables, and the attribute parameters of the preset solution variables correspond to the basic attribute parameters and the preset attribute parameters of at least one dimension. The target production scheduling model construction module is used to construct a target production scheduling model based on the sample production scheduling data and the corresponding sample production scheduling results, which is used to output the production scheduling results that characterize the production scheduling optimization scheme. The target production scheduling model construction module includes: The first production scheduling model construction unit is used to construct a first production scheduling model for the corresponding production scheduling problem based on the sample production scheduling data and the corresponding sample production scheduling results. An initial production scheduling model construction unit is used to obtain an initial production scheduling model based on the first production scheduling model for different production scheduling problems. The first production scheduling result acquisition unit is used to use the solver to solve the sample production scheduling data and obtain the corresponding first production scheduling result. The target production scheduling module acquisition unit is used to retrain the first production scheduling model based on the sample production scheduling data and the first production scheduling result to optimize the initial production scheduling model until a preset optimization condition is met to obtain the final target production scheduling model.

10. The production scheduling model building apparatus as described in claim 9, wherein the building apparatus further comprises: The parameter adjustment instruction acquisition module is used to acquire attribute parameter adjustment instructions; The attribute parameter adjustment module is used to set other attribute parameters of the variables in the solver besides the basic attribute parameters, based on the attribute parameter adjustment instructions, using at least one of the adjustment methods of modification, addition, and deletion.

11. The production scheduling model construction apparatus as described in claim 9, wherein the first production scheduling result acquisition unit is used to use the solver to distinguish and process the preset solution variables in the sample production scheduling data according to their priorities, so as to obtain the first production scheduling result corresponding to the preset solution variables of different priorities in the sample production scheduling data respectively.

12. The production scheduling model construction device as described in claim 9, wherein the sample production scheduling data acquisition module is used to acquire several sets of sample production scheduling data corresponding to each production scheduling problem in the flexible work workshop using a bipartite diagram; The sample production scheduling result acquisition module is used to obtain the corresponding sample production scheduling result based on the production scheduling result identification information corresponding to the sample production scheduling data.

13. The production scheduling model construction apparatus as described in claim 9, wherein the initial production scheduling model construction unit is used to perform fusion processing on the first production scheduling models of different production scheduling problems using a preset fusion method to obtain the initial production scheduling model.

14. A production scheduling device for a flexible workshop, wherein the production scheduling method is implemented based on a target production scheduling model obtained by the production scheduling model construction device according to any one of claims 9-13, the production scheduling device comprising: The actual production scheduling data acquisition module is used to collect several sets of actual production scheduling data corresponding to different scheduling problems in each scheduling task in the flexible operation workshop. The production scheduling result prediction module is used to input each set of actual production scheduling data into the target production scheduling model to obtain the output predicted production scheduling result; The solution result acquisition module is used to solve the target production scheduling model using a solver to obtain the solution result; The fixed variable determination module is used to determine the corresponding preset solution variables included in each group of actual production data as fixed variables based on the predicted production scheduling results and the solution results. Other solution variable acquisition module is used to acquire other solution variables besides the fixed variables among all the preset solution variables corresponding to each group of actual production scheduling data; The target production scheduling result acquisition module is used to use the solver to solve the other solution variables to obtain the target production scheduling result for each group of actual production scheduling data.

15. The production scheduling device for a flexible workshop as described in claim 14, wherein the fixed variable determination module comprises: The first hybridization result acquisition unit is used to perform hybridization processing on the predicted production scheduling result and the solution result when the solution result characterizes the existence of a feasible solution, and to obtain the first hybridization result. The first fixed variable determination unit is used to determine the corresponding preset solution variable included in each group of actual production data as the fixed variable based on the first hybridization result.

16. The production scheduling device for a flexible workshop as described in claim 15, wherein the fixed variable determination module comprises: The second hybridization result acquisition unit is used to acquire the linear programming LP solution in the solver when the solution result indicates that there is no feasible solution, and to perform hybridization processing on the predicted production scheduling result and the linear programming LP solution to obtain the second hybridization result; The second fixed variable determination unit is used to determine the corresponding preset solution variable included in each group of actual production data as the fixed variable based on the second hybridization result.

17. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method of any one of claims 1-5; or, the method of any one of claims 6-8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-5; or, the method according to any one of claims 6-8.

19. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-5; or, the method according to any one of claims 6-8.

Images