Optimisation system, optimisation method, and optimisation program
The optimization system addresses large-scale job shop scheduling challenges by dividing jobs into ordered groups and iteratively optimizing schedules, achieving near-optimal solutions with high accuracy and stability.
Patent Information
- Application Number
- PCT/JP2026/017415
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-16
- Filing Date
- 2026-05-01
- Publication Date
- 2026-11-19
AI Technical Summary
Solving large-scale job shop scheduling problems with complex constraints requires significant computational resources and is challenging due to the difficulty in integrating partitioned subproblems while maintaining overall consistency.
An optimization system that divides jobs into ordered groups, generates an initial solution, and iteratively adjusts schedules using objective functions and constraints to find a near-optimal solution, incorporating methods like integer programming and constraint programming to handle complex constraints.
Efficiently finds near-optimal solutions to job shop scheduling problems, even for large-scale scenarios with complex constraints, ensuring high accuracy and stability while maintaining overall consistency.
Smart Images

Figure JP2026017415_19112026_PF_FP_ABST
Abstract
Description
Optimization system, optimization method, and optimization program
[0001] The present disclosure relates to an optimization system, an optimization method, and an optimization program.
[0002] Japanese Unexamined Patent Application Publication No. 2007-102435 describes a mode in which, based on predetermined schedule objectives and operational conditions, an initial solution for the job processing order is generated, a plurality of jobs from the start of the processing order are extracted as partial plan target jobs from among the jobs whose processing order in the initial solution is undecided, the processing order of the partial plan target jobs that gives the optimal evaluation value for all permutations of the partial plan target jobs is selected, one or more jobs from the start of the selected processing order of the partial plan target jobs are determined as the relevant processing order, and the processes of acquiring partial plan target jobs, evaluating the processing order, performing partial plan optimization calculation, and determining the partial plan processing order are repeated until the processing order of all jobs is determined.
[0003] Japanese Patent No. 7464307 describes a mode in which the problem of determining the machining order of a plurality of workpieces is formulated as a traveling salesman problem that minimizes the sum of setup change times associated with workpiece switching, while satisfying delivery date constraints, order constraints defined between the plurality of workpieces, and order constraints defined between groups including the plurality of workpieces, the solution to the job shop problem is calculated using an annealing processing means, and the machining order of the groups is determined based on the calculation result.
[0004] Japanese Unexamined Patent Application Publication No. 2024-118657 describes a mode in which a first plan related to a first task and a second plan related to a second task are evaluated, the first constraint condition of the first plan creation unit is modified according to the evaluation result, and the first plan is updated based on the adjusted first constraint condition.
[0005] Japanese Patent Publication No. 2007-222911 describes a system for determining the arrangement order of multiple processing steps for processing multiple steel plates, wherein multiple processing steps whose attribute values of a first attribute are equal to each other are classified into groups, and for each of the multiple processing steps classified into a certain group, the multiple processing steps are rearranged in an order that satisfies a second order constraint and a third order constraint, a portion of the rearranged multiple processing steps is selected as a cluster, and the multiple processing steps are arranged in a manner that maximizes the processing efficiency of the steel plates by solving an integer programming problem based on the constraints between the multiple clusters.
[0006] This disclosure is made in view of the above circumstances and aims to provide an optimization system, optimization method, and optimization program that can efficiently find a solution to the job shop scheduling problem.
[0007] The first embodiment of the optimization system includes a processor and is an optimization system that solves a job shop scheduling problem for multiple jobs, wherein the processor generates an initial solution to the original problem, which is the job shop scheduling problem to be solved, selects an arbitrary group of jobs from the group of jobs obtained when generating the initial solution, and repeatedly adjusts the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that do not worsen the value of the objective function, thereby obtaining a near-optimal solution to the original problem.
[0008] The optimization system of the second embodiment generates the initial solution by, in the optimization system of the first embodiment, the processor divides the plurality of jobs into a plurality of ordered groups when generating the initial solution, selects one group at a time from the beginning of the plurality of groups, and creates a schedule for the jobs included in the selected group, and performs the same for all groups when creating a schedule for the jobs included in the selected group, and performs optimization under the objective function and constraints of the original problem and constraints that fix the decision variables related to the jobs included in the group for which a schedule has already been created.
[0009] The third embodiment of the optimization system is an optimization system in which, in the second embodiment, the processor divides the plurality of jobs into a plurality of ordered groups, taking into consideration the constraints on the order between jobs included in the original problem.
[0010] The optimization system of the fourth embodiment is an optimization system of the third embodiment in which, when the processor divides the plurality of jobs into a plurality of ordered groups, it divides the plurality of jobs into a plurality of ordered groups based on constraints on the order between jobs included in the original problem, such that later jobs belong to groups following the group to which earlier jobs belong.
[0011] The fifth aspect of the optimization method is an optimization method for solving a job shop scheduling problem for multiple jobs, wherein the processor of the information processing device executes a process to find a near-optimal solution to the original problem, which is the job shop scheduling problem to be solved, by generating an initial solution for the original problem, which is the job shop scheduling problem to be solved, selecting an arbitrary group of jobs from the group of jobs obtained when generating the initial solution, and repeatedly adjusting the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that do not worsen the value of the objective function.
[0012] The sixth embodiment of the optimization program is an optimization program for solving a job shop scheduling problem for multiple jobs, which involves generating an initial solution to the original problem, which is the job shop scheduling problem to be solved; selecting an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjusting the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby causing a processor in the information processing device to execute a process to find a near-optimal solution to the original problem.
[0013] According to this disclosure, it is possible to efficiently find a solution to the job shop scheduling problem.
[0014] This is a block diagram showing an example of the schematic configuration of an optimization system. This is a block diagram showing an example of the server hardware configuration. This is a block diagram showing an example of the server's functional configuration. This is a block diagram showing an example of the server's initial solution generation unit's functional configuration. This is a flowchart showing an example of the optimization process. This is a diagram illustrating an example of the job shop scheduling problem. This is a diagram illustrating the processing of each step of the optimization process. This is a diagram showing an example of product type data in an embodiment. This is a diagram showing an example of work sequence data in an embodiment. This is a diagram showing an example of assignment feasibility data in an embodiment. This is a diagram showing an example of inter-product constraint data in an embodiment. This is a diagram showing an example of the correspondence between product types, work sets, and tasks in an embodiment. This is a diagram illustrating the process of selecting multiple product types and adjusting the plan in an embodiment.
[0015] Hereinafter, with reference to the drawings, examples of embodiments for carrying out the technology of this disclosure will be described in detail.
[0016] <Overview of this Embodiment> The job shop scheduling problem is an optimization problem that aims to efficiently schedule multiple jobs and their associated operations (tasks) on limited resources. Solving large-scale problems all at once requires an enormous amount of computation time, making it difficult to generate practical schedules. Problem partitioning is widely used as a way to deal with such large-scale problems, but especially in large-scale problems with complex constraints, there is a challenge in integrating the partitioned subproblems while maintaining overall consistency.
[0017] Therefore, this embodiment aims to efficiently find a solution to the job shop scheduling problem.
[0018] <System Configuration> Figure 1 is an explanatory diagram showing an example configuration of the optimization system 2. In this embodiment, the optimization system 2 that provides a near-optimal solution to the job shop scheduling problem to the user through optimization will be described. The optimization system 2 includes a server 10 and a user terminal 90. The server 10 and the user terminal 90 are connected via a network N.
[0019] Server 10 is, for example, a server computer capable of various information processing and information transmission / reception. Note that the device corresponding to Server 10 is not limited to a server computer; for example, it may be a personal computer or the like. In this embodiment, Server 10 functions as a device that provides the user with a near-optimal solution to the job shop scheduling problem through optimization.
[0020] The user terminal 90 is a general-purpose computer such as a personal computer. In this embodiment, the user terminal 90 functions as a device that presents the user with a near-optimal solution to the job shop scheduling problem.
[0021] In this embodiment, the server 10 provides the user with a near-optimal solution to the job shop scheduling problem through optimization. However, the local user terminal 90 may also provide the user with a near-optimal solution to the job shop scheduling problem through optimization. In other words, the distinction between the two is merely for convenience, and a single computer may perform the entire process.
[0022] <Server Configuration According to This Embodiment> Referring to Figure 2, the hardware configuration of the server 10 according to this embodiment will be described. An example of the server 10 is a personal computer or a server computer. As shown in Figure 1, the server 10 includes a CPU (Central Processing Unit) 20, a memory 21 as a temporary storage area, and a non-volatile storage unit 22. The server 10 also includes a display 23 such as a liquid crystal display, input devices 24 such as a keyboard and mouse, and a network interface 25 connected to a network. The CPU 20, memory 21, storage unit 22, display 23, input devices 24, and network interface 25 are connected to a bus 27. The CPU 20 is an example of a processor according to the disclosed technology.
[0023] The storage unit 22 is implemented by an HDD (Hard Disk Drive), SSD (Solid State Drive), or flash memory, etc. The storage unit 22, as a storage medium, stores the information processing program 30. The CPU 20 reads the information processing program 30 from the storage unit 22, expands it into the memory 21, and executes the expanded information processing program 30. The information processing program 30 is an example of an optimization program.
[0024] Furthermore, the memory unit 22 stores problem data 32, which is a problem example of a job shop scheduling problem that the server 10 aims to solve, in a data format that the server 10 can process. The problem data 32 includes a set of elements that constitute the job shop scheduling problem to be solved, and includes a set of elements to be solved that consists of multiple elements. The problem data 32 also includes constraints.
[0025] The elements that make up the set of elements to be solved refer to, for example, in the case of a job shop scheduling problem, the worker, the work date and time, and the type of product being worked on. Specific examples of the set of elements to be solved will be explained in the examples described later.
[0026] Incidentally, the server 10 according to this embodiment has the function of performing optimization calculations on an example of a job shop scheduling problem to be solved, and presenting a near-optimal solution to the job shop scheduling problem to the user.
[0027] Next, with reference to Figure 3, the functional configuration of the server 10 according to this embodiment will be described. As shown in Figure 3, the server 10 includes an initial solution generation unit 40, a selection unit 42, an adjustment unit 44, an iteration determination unit 46, and an output unit 48. The CPU 20 executes the information processing program 30, thereby enabling the initial solution generation unit 40, the selection unit 42, the adjustment unit 44, the iteration determination unit 46, and the output unit 48 to function.
[0028] The initial solution generation unit 40 generates an initial solution for the original problem, which is a job shop scheduling problem to be solved.
[0029] Specifically, as shown in Figure 4, the initial solution generation unit 40 includes a division unit 40A and a creation unit 40B.
[0030] The division unit 40A divides the multiple jobs into multiple ordered groups. Specifically, when the division unit 40A divides the multiple jobs into multiple ordered groups, it takes into account the constraints on the order between jobs included in the original problem, and divides the multiple jobs into multiple ordered groups.
[0031] More specifically, when the division unit 40A divides multiple jobs into multiple ordered groups, it performs the following processing: That is, based on the constraints on the order between jobs included in the original problem, the division unit 40A divides multiple jobs into multiple ordered groups such that later jobs belong to groups following the group to which earlier jobs belong.
[0032] The creation unit 40B generates an initial solution by selecting one group at a time from the beginning of multiple groups and creating a schedule for the jobs included in the selected group, and then performing this process for all groups.
[0033] When the creation unit 40B creates a schedule for the jobs included in the selected group, it performs optimization under the objective function and constraints of the original problem, and constraints that fix the decision variables related to the jobs included in the group for which a schedule has already been created. At this time, the creation unit 40B uses software called a solver to solve the optimization problem for creating a schedule for the jobs included in the selected group.
[0034] The selection unit 42 selects an arbitrary group of jobs from a plurality of job groups obtained when generating the initial solution. For example, it selects an arbitrary group of jobs from a group of jobs arranged based on the results of the initial solution.
[0035] The adjustment unit 44 adjusts the schedule for the selected group of jobs so that they are optimized under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating.
[0036] The repetition determination unit 46 determines whether or not a predetermined repetition termination condition is met, and causes the selection unit 42 and the adjustment unit 44 to repeat their respective processes until it is determined that the repetition termination condition is met.
[0037] Examples of the conditions for terminating the iteration mentioned above include the condition that the value of the objective function has converged, or the condition that the number of iterations exceeds the upper limit.
[0038] The output unit 48 outputs the final adjusted schedule as a near-optimal solution to the original problem.
[0039] Next, the operation of the server 10 according to the present embodiment will be described with reference to FIG. 5. The optimization processing shown in FIG. 5 is executed by the CPU 20 executing the information processing program 30. The optimization processing shown in FIG. 5 is executed, for example, when an execution start instruction is input by a user.
[0040] In step S100 of FIG. 5, the dividing unit 40A divides a plurality of jobs into a plurality of ordered groups.
[0041] At this time, the plurality of jobs are divided into a plurality of ordered groups in consideration of inter-job constraints as shown in FIG. 6 (see STEP 1 in FIG. 7). FIG. 6 shows an example in which there is an inter-job constraint between job 1 and job 2. Further, STEP 1 in FIG. 7 shows an example of dividing the jobs into a plurality of groups such that job 2 and job 3 belong to group (1), and job 1 and job 4 belong to group (2). That is, in this example, since there is an inter-job constraint between job 1 and job 2, the plurality of jobs are divided into a plurality of ordered groups such that job 2 belongs to group 2, which is subsequent to group 1 to which job 1 belongs.
[0042] In step S102, the creating unit 40B sequentially selects one group from the plurality of groups.
[0043] In step S104, the creating unit 40B creates a schedule for the jobs included in the selected group (see STEP 2 in FIG. 7). STEP 1 in FIG. 7 shows an example in which after creating schedules for job 2 and job 3 belonging to group (1), schedules are created for job 1 and job 4 belonging to group (2).
[0044] In step S106, the initial solution generation unit 40 determines whether or not the processing of steps S102 to S104 described above has been executed for all of the plurality of groups. When the processing of steps S102 to S104 described above has been executed for all of the plurality of groups, the process proceeds to step S108. On the other hand, when there is a group for which the processing of steps S102 to S104 described above has not been executed, the process returns to step S102, and the processing is repeated with the relevant group as the target.
[0045] In step S108, the selection unit 42 selects an arbitrary job group from a plurality of job groups obtained when generating the initial solution (see STEP 3 in FIG. 7). STEP 3 in FIG. 7 shows an example in which a job group consisting of job 2, job 3, and job 1 is selected first, and then a job group consisting of job 2, job 3, and job 4 is selected.
[0046] In step S110, the adjustment unit 44 adjusts the schedule so as to perform optimization for the selected job group under the objective function and constraint conditions of the original problem, the constraint condition for fixing decision variables related to unselected job groups, and the constraint condition that the value of the objective function does not deteriorate (see STEP 3 in FIG. 7). STEP 3 in FIG. 7 shows an example in which the schedule is adjusted first for the job group consisting of job 2, job 3, and job 1, and then the schedule is adjusted for the job group consisting of job 2, job 3, and job 4.
[0047] In step S112, the iteration determination unit 46 determines whether or not a predetermined iteration termination condition is satisfied. If it is determined that the iteration termination condition is satisfied, the process proceeds to the above step S114. If it is determined that the iteration termination condition is not satisfied, the process returns to the above step S108.
[0048] In step S114, the output unit 48 displays the finally adjusted schedule on the user terminal 90 as a suboptimal solution of the original problem, and terminates the optimization processing.
[0049] Next, examples will be described.
[0050] <Example> In this example, we will describe an example in which the disclosed technology is applied to the problem of minimizing the last completed operation (hereinafter referred to as "makespan"). Specifically, the problem to be solved is an optimization problem in which, within a certain period, all operations are planned to be completed as quickly as possible by efficiently assigning 20 types of products to coating equipment one type at a time. In the coating process, each type of product must go through specific operations such as priming, intermediate coating, top coating, finishing, and inspection, and each is assigned to a dedicated or shared coating machine. Each operation takes a certain amount of time depending on the characteristics of the product type, and time constraints (time difference constraints between operations) are strictly defined. In addition, at the end of this period, the equipment is inspected and maintained, but until then, a constraint is imposed that a specific type of product must be processed by a specific machine only after the processing of other specific types of products has been completed (this will be called an inter-product constraint).
[0051] The problem data 32 relating to the embodiment includes the variety data shown in Figure 8. As shown in Figure 8, the variety data associates the start time of work for each variety to be manufactured with the delivery time for that variety.
[0052] Furthermore, the problem data 32 relating to the embodiment includes the work sequence data shown in Figure 9. As shown in Figure 9, the work sequence data includes, for each product type, the work required for manufacturing that product type, constraints on the work sequence, the time required for each work, and the waiting time required after each work. In the example in Figure 9, the arrows represent constraints on the work sequence, the time indicated near the arrows represents the waiting time after the work is completed, and the time indicated to the right of the work represents the time required for each work. Also, in the example in Figure 9, product type 1 is shown to require a waiting time of 30 minutes or more after the undercoat is completed before the intermediate coat is applied, and a waiting time of 0 minutes or more after the intermediate coat is applied before the top coat is applied. Also, in the example in Figure 9, product type 1 is shown to require 8 minutes for the undercoat, 5 minutes for the intermediate coat, and 3 minutes for the top coat.
[0053] Furthermore, in the example in Figure 9, variety 2 is shown to have a waiting period of 0 minutes or more after the undercoat is completed before the intermediate coat is applied, and a waiting period of 0 minutes or more after the intermediate coat is applied before the topcoat is applied. Furthermore, in the example in Figure 9, variety 2 is shown to have a waiting period of 5 minutes or more after the undercoat is completed before the inspection is performed, and a waiting period of 0 minutes or more after the inspection before the topcoat is applied. Furthermore, in the example in Figure 9, variety 2 is shown to have a required time of 7 minutes for the undercoat, a required time of 10 minutes for the intermediate coat, a required time of 3 minutes for the inspection, and a required time of 5 minutes for the topcoat.
[0054] Furthermore, the problem data 32 relating to the embodiment includes data on assignment feasibility as shown in Figure 10. As shown in Figure 10, the data on assignment feasibility includes information indicating whether the coating device can be assigned to a given task for each combination of work and coating device for each product type. In the example in Figure 10, a circle indicates that the coating device can be assigned to that task for that product type, and a blank space indicates that the coating device cannot be assigned to that task for that product type. In the example in Figure 10, coating device A can be assigned to the undercoat for product type 1, but not to the intermediate coat and topcoat, and for product type 2, it can be assigned to the undercoat and topcoat, but not to the intermediate coat and inspection.
[0055] Furthermore, as shown in Figure 11, the problem data 32 relating to the embodiment includes data on inter-variety constraints as shown in Figure 11. As shown in Figure 11, the data on inter-variety constraints includes information representing constraints on the order between varieties.
[0056] In the example in Figure 11, the column labeled "Preceding" indicates a constraint that processing of the "Subsequent" variety must be carried out only after processing of the "Preceding" variety is completed, while a blank column indicates no constraint on the order between the "Preceding" and "Subsequent" varieties. In the example in Figure 11, there is a constraint that processing of variety 2 must be carried out only after processing of variety 1 is completed.
[0057] Furthermore, as shown in Figure 12, in this embodiment, a combination of variety and work is treated as a task. In the example in Figure 12, the combination of variety 1 and undercoating is shown as task 1.
[0058] In the examples, the above data will be represented by the symbols shown below.
[0059]
[0060] Furthermore, the problem data 32 relating to the embodiment includes decision variables, constraints, and an objective function in order to formulate it as a mixed-integer linear programming problem. The decision variables are as follows: i and a i,w It is defined as follows.
[0061]
[0062] The constraints are as follows: (1) to (7).
[0063] Constraint (1): Each task must begin after the start time. Constraint (2): Each task must be completed by the deadline. Constraint (3): Each task must be assigned to one coating device. Constraint (4): Only one task can be assigned to each coating device at a time. Constraint (5): The order between tasks must be maintained. Constraint (6): Each task must only be assigned to a coating device that is available for assignment. Constraint (7): Inter-product constraints must be observed.
[0064] The objective function is to complete the last task as early as possible.
[0065] In this embodiment, the division unit 40A divides the varieties (=jobs) into multiple sequential groups. Here, 20 varieties are divided into 5 varieties x 4 groups.
[0066] The following constants are defined.
[0067]
[0068] The data is grouped using an arbitrary method. This grouping is formulated as an integer programming problem as follows. The decision variable is x, as shown below. hg It is defined as follows.
[0069]
[0070] The constraints are shown by equations (1) to (3) below. Equation (1) represents the condition that each variety is assigned to one group. Equation (2) represents the condition that the number of varieties assigned to each group is maintained. Equation (3) represents the condition that inter-varietal constraints are maintained.
[0071] ... (1) ... (2) ... (3)
[0072] The objective function is defined by the following equation, which is to assign lots with earlier delivery dates to the earliest possible group.
[0073]
[0074] The division unit 40A groups multiple varieties by solving the above integer programming problem using an arbitrary solver.
[0075] The creation unit 40B creates schedules sequentially from the first group divided by the division unit 40A. General job shop scheduling solutions (branch and bound method, constraint programming, genetic algorithms, simulated annealing, tabu search, etc.) can be used to create the schedules. Here, in order to formulate it as an integer programming problem, constants, decision variables, constraints, and the objective function are defined as follows.
[0076] Constants are defined as follows:
[0077]
[0078] The decision variable is shown below: s i and a iw It is defined as follows.
[0079]
[0080] The constraints are shown in equations (4) to (12) below. Equation (4) represents the condition that each task starts after the start time. Equation (5) represents the condition that each task is completed by the deadline. Equation (6) represents the condition that each task is assigned to one coating device. Equation (7) represents the condition that only one task can be assigned to each coating device. Equation (8) represents the condition that the order between tasks is maintained. Equation (9) represents the condition that each task is assigned only to a coating device that can be assigned to it. Equation (10) represents the condition that inter-product constraints are maintained. Equation (11) represents the condition that the start time of already scheduled tasks is fixed. Equation (12) represents the condition that the assignment variables of already scheduled tasks are fixed.
[0081] ... (4) ... (5) ... (6) ... (7) ... (8) ... (9) ... (10) ... (11) ... (12)
[0082] The objective function is to complete the last task to be completed as early as possible, and is shown by equation (13) below.
[0083] ... (13)
[0084] The creation unit 40B solves the above integer programming problem for all tasks in order from group g=1 using an arbitrary solver. Once the processing for group g=4 is complete, an initial schedule for the original problem is obtained.
[0085] The selection unit 42 selects multiple varieties as an arbitrary job group from the initial schedule obtained above, and the adjustment unit 44 adjusts the plan. This process is repeated multiple times. Here, seven varieties are selected from the 20 varieties as an arbitrary job group. The varieties to be selected are those that are "1st to 7th", "4th to 10th", "7th to 13th", etc., when the varieties are sorted in makespan order based on the obtained schedule, with overlaps allowed and the selections slightly shifted, and the schedule is adjusted for the seven varieties (Figure 13).
[0086] Common job shop scheduling methods (branch and bound, constraint programming, genetic algorithms, simulated annealing, taboo search, etc.) can be used to adjust the schedule. Here, we define constants, decision variables, constraints, and an objective function to formulate it as an integer programming problem as follows.
[0087] Constants are defined as follows:
[0088]
[0089] The decision variable is shown below: s i and a iw It is defined as follows.
[0090]
[0091] The constraints are shown in equations (14) to (23) below. Equation (14) represents the condition that each task starts after the start time. Equation (15) represents the condition that each task is completed by the deadline. Equation (16) represents the condition that each task is assigned to one coating device. Equation (17) represents the condition that only one task can be assigned to each coating device. Equation (18) represents the condition that the order between tasks is maintained. Equation (19) represents the condition that each task is assigned only to a coating device that can be assigned to it. Equation (20) represents the condition that inter-product constraints are maintained. Equation (21) represents the condition that the start time of unadjusted tasks is fixed. Equation (22) represents the condition that the assignment variables of unadjusted tasks are fixed. Equation (23) represents the condition that the objective function is not worsened from before the adjustment.
[0092] ... (14) ... (15) ... (16) ... (17) ... (18) ... (19) ... (20) ... (21) ... (22) ... (23)
[0093] The objective function is to complete the last task to be completed as early as possible, and is shown by equation (24) below.
[0094] ... (24)
[0095] The adjustment unit 44 obtains a near-optimal solution to the original problem by repeatedly solving the above integer programming problem for each of the seven selected varieties.
[0096] As described above, this embodiment makes it possible to efficiently find a solution to the job shop scheduling problem.
[0097] Furthermore, in job shop scheduling problems, even for complex problems involving constraints between operations (such as sequence constraints and time difference constraints) and constraints between jobs, the system can partition the problem in a way that prevents it from becoming unexecutable, and can stably and accurately find solutions even for large-scale problems.
[0098] Furthermore, it is possible to find a near-optimal solution with high accuracy and speed based on problem decomposition that takes complex constraints into account.
[0099] Furthermore, when integrating the solutions of the divided subproblems, it is possible to stably obtain highly accurate near-optimal solutions while maintaining the overall consistency.
[0100] Furthermore, because it can handle large-scale problems, it can significantly expand the range of industrial applications.
[0101] In the above embodiment, for example, the hardware structure of a processing unit that performs various processes, such as each functional unit of the server 10, can be the following types of processors. As mentioned above, these types of processors include a CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, as well as a programmable logic device (PLD), such as an FPGA (Field Programmable Gate Array), whose circuit configuration can be changed after manufacturing, and a dedicated electrical circuit, such as an Application Specific Integrated Circuit, which has a circuit configuration specifically designed to perform a particular process.
[0102] A single processing unit may be composed of one of these various processors, or it may be composed of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, multiple processing units may be composed of a single processor.
[0103] Examples of configuring multiple processing units with a single processor include, firstly, a configuration where one or more CPUs and software combine to form a single processor, as exemplified by client and server computers, and this processor functions as multiple processing units. Secondly, a configuration using a processor that realizes the functions of the entire system, including multiple processing units, on a single IC (Integrated Circuit) chip, as exemplified by System-on-a-Chip (SoC). Thus, various processing units are configured, in terms of hardware structure, using one or more of the above-mentioned processors.
[0104] Furthermore, the hardware structure of these various processors can more specifically utilize an electrical circuit (circuitry) that combines circuit elements such as semiconductor elements.
[0105] Furthermore, although the above embodiment describes a configuration in which the information processing program 30 is pre-stored (installed) in the storage unit 22, the invention is not limited to this configuration. The information processing program 30 may be provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the information processing program 30 may be provided in a form that is downloaded from an external device via a network.
[0106] The program of this application can be provided as a program product. A program product includes all forms of products for providing a program. For example, a program product includes a program provided via a network such as the Internet, and non-temporary computer-readable recording media such as CD-ROMs and DVDs on which the program is stored.
[0107] The following additional information is disclosed regarding the embodiments described above.
[0108] (Note 1) An optimization system that includes a processor and solves a job shop scheduling problem for multiple jobs, wherein the processor generates an initial solution to the original problem, which is the job shop scheduling problem to be solved; selects an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjusts the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby finding a near-optimal solution to the original problem.
[0109] (Note 2) The optimization system according to Note 1, wherein the processor generates the initial solution by dividing the plurality of jobs into a plurality of ordered groups, selecting one group at a time from the beginning of the plurality of groups, and creating a schedule for the jobs included in the selected group, and when creating a schedule for the jobs included in the selected group, performs optimization under the objective function and constraints of the original problem and constraints that fix the decision variables related to the jobs included in the group for which a schedule has already been created.
[0110] (Note 3) The processor is an optimization system according to Note 2, which divides the plurality of jobs into a plurality of ordered groups, taking into consideration the constraints on the order between jobs included in the original problem.
[0111] (Note 4) The optimization system described in Note 3, wherein the processor divides the plurality of jobs into a plurality of ordered groups, and divides the plurality of jobs into a plurality of ordered groups based on constraints on the order between jobs included in the original problem, such that later jobs belong to groups following the group to which earlier jobs belong.
[0112] (Note 5) An optimization method for solving a job shop scheduling problem for multiple jobs, wherein the processor of the information processing device performs the following steps: generate an initial solution for the original problem, which is the job shop scheduling problem to be solved; select an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjust the schedule of the selected group of jobs so as to perform optimization under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby finding a suboptimal solution to the original problem.
[0113] (Note 6) An optimization program for solving a job shop scheduling problem for multiple jobs, wherein the program generates an initial solution to the original problem, which is the job shop scheduling problem to be solved; selects an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjusts the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby causing the processor of the information processing device to execute a process to find a near-optimal solution to the original problem.
[0114] The disclosure of Japanese Patent Application 2025-082648 is incorporated herein by reference in its entirety.
[0115] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually described as being incorporated by reference.
Claims
1. An optimization system including a processor that solves a job shop scheduling problem for multiple jobs, wherein the processor generates an initial solution to the original problem, which is the job shop scheduling problem to be solved; selects an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjusts the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby finding a near-optimal solution to the original problem.
2. The optimization system according to claim 1, wherein the processor generates the initial solution by dividing the plurality of jobs into a plurality of ordered groups, selecting one group at a time from the beginning of the plurality of groups, and creating a schedule for the jobs included in the selected group, and performing this for all groups, and when creating a schedule for the jobs included in the selected group, performing optimization under the objective function and constraints of the original problem and constraints that fix the decision variables related to the jobs included in the group for which a schedule has already been created.
3. The optimization system according to claim 2, wherein the processor divides the plurality of jobs into a plurality of ordered groups, taking into consideration the constraints relating to the order between jobs included in the original problem.
4. The optimization system according to claim 3, wherein the processor divides the plurality of jobs into a plurality of ordered groups, and divides the plurality of jobs into a plurality of ordered groups based on constraints on the order between jobs included in the original problem, such that later jobs belong to groups following the group to which earlier jobs belong.
5. An optimization method for solving a job shop scheduling problem for multiple jobs, wherein the processor of an information processing device performs the following steps: generate an initial solution to the original problem, which is the job shop scheduling problem to be solved; select an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjust the schedule of the selected group of jobs so as to perform optimization under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby finding a near-optimal solution to the original problem.
6. An optimization program for solving a job shop scheduling problem for multiple jobs, wherein the program generates an initial solution to the original problem, which is the job shop scheduling problem to be solved; selects an arbitrary group of jobs from the group of jobs obtained when generating the initial solution; and repeatedly adjusts the schedule to optimize the selected group of jobs under the objective function and constraints of the original problem, constraints that fix the decision variables related to the unselected group of jobs, and constraints that prevent the value of the objective function from deteriorating, thereby causing a processor in an information processing device to execute a process to find a near-optimal solution to the original problem.
