JP5318651B2 - Scheduling method, scheduling program, and scheduling system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To find an optimized planned production schedule, even when a rescheduling is required in real time because of operation fluctuation. <P>SOLUTION: The optimum solution is drawn out in a pre-given initial scheduling problem, to be stored as a planned production schedule in a storage part S1. a process state after a fixed time is predicted based on the planned production schedule S3, in a time point after starting an operation S2. A rescheduling problem required to be solved when the operation fluctuation estimated to occur occurs is prepared in the predicted process, until the whole production is finished S4, and a solution thereof is drawn out, to be stored into the storage part as the estimated rescheduling problem optimum solution S5, S6. Then, the presence of generation of the preliminarily recorded operation fluctuation estimated to occur as an estimated operation fluctuation is determined S8 after the fixed time lapses S7, and the rescheduling problem optimum solution stored in the storage part is changed S9 into the planned production schedule hereinafter, when the operation fluctuation estimated to occur occurs. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a scheduling method, a scheduling program, and a scheduling system for rescheduling a production schedule.

  Normally, the production schedule for a factory needs to be considered in terms of productivity, delivery date, cost, etc., and is created so as to realize the optimization. In particular, scheduling of a production schedule at a material production site is a complicated combination optimization problem, and a long calculation time is often required to obtain an optimal production schedule by a computer. However, when the production schedule for one day is derived before the start of the day, the length of calculation time does not matter much. For example, it is assumed that one hour is required for calculation for deriving an optimal production schedule. In this case, if the calculation is started one hour before the operation start time, an optimum production schedule can be derived before the operation starts.

  On the other hand, in the rescheduling problem, this calculation time becomes a big problem. The rescheduling problem is a problem in which a production schedule is changed when a change occurs in a target process (for example, trouble or quality failure in equipment) or a change occurs in another process. This is necessary because the original production schedule cannot be dealt with due to fluctuations, or a better production schedule can be created than the original production schedule.

  In the rescheduling problem, it is desirable to respond in real time by changing the production schedule as soon as fluctuation occurs. This makes it unacceptable to wait for one hour for optimization calculations. However, for the rescheduling problem, it is not a realistic solution to obtain an optimal solution using an optimization algorithm when the problem occurs because it takes too much computation time. Therefore, conventionally, with respect to the rescheduling problem, a technique for performing rescheduling manually by a person in charge of scheduling or performing rescheduling by a rule-based system that requires a short calculation time is used. For example, in the technology disclosed in Patent Document 1, in the rescheduling process, among the orders in which it is difficult to finish the product within the delivery date, the working time of the machine is increased for an order having a high operation rate of the processing machine to be used. For orders where the operating rate of the manufacturing machine to be used is low, such an order is deleted, and the execution of the work is postponed, so that the idle state when the machine is operating is not created. Based on this, a machining schedule is drafted. In addition, the technology disclosed in Patent Document 2 is based on a rule that preferentially uses reserved parts stored as intermediate stock in rescheduling processing, thereby reducing useless intermediate inventory and reducing lead time. The plan to plan is made. However, in any case, compared to an optimized production schedule, solutions that are inferior in terms of productivity, delivery time, cost, etc. were to be used.

JP-A-6-44258 JP-A-6-203037

  A problem to be solved by the present invention is a scheduling method, a scheduling program, and a scheduling system for obtaining an optimized production schedule even when rescheduling is required in real time due to operational fluctuations of a process to be scheduled or upper and lower processes. Is to provide.

According to the scheduling method of the present invention, in a computer having a calculation unit, a storage unit, an input unit, and an output unit, a process executed by the calculation unit is manufactured in a target process given in advance at the beginning of a day. that it the initial scheduling of the given manufacturing candidate as the type and number upper limit of the product being allowed scheduling derive an optimal solution for determining the production order in the product type and number as well as the target step optimum A scheduled production schedule recording step for recording an optimal solution derived for the initial scheduling problem in the storage unit as a scheduled production schedule, and at a certain time after the start of operation, after a certain time in the scheduled production schedule A process state prediction step for predicting the process state from the scheduled production schedule, The virtual is being operated variation, the expected operating fluctuations recording step of recording in the storage unit as the expected operating fluctuations, in the process state after the step state said predetermined time predicted by the prediction step, for said manufacture candidate have not yet produced undertaken A rescheduling problem prediction step that creates a rescheduling problem that needs to be solved in response to the expected operational fluctuation when the expected operational fluctuation recorded as the predicted operational fluctuation occurs; For the rescheduling problem, an optimal solution that determines the type and number of products and the manufacturing order in the target process is derived, and the operation that is expected to occur as the optimal solution for the rescheduling problem that is predicted and rescheduling problem optimal solution created recording step of recording in the storage unit in response to the variation, the one After time, the expected when operating recording the generated operation expected variations were as variation actually occurs, the rescheduling problem optimal solution the generating recorded in the storage unit corresponding to the operation expected variations A scheduled production schedule change step for overwriting the scheduled production schedule recorded in the storage unit, the process state prediction step, the rescheduling problem prediction step, the rescheduling problem optimal solution creation recording step, It is characterized by being repeatedly executed at regular intervals.

The scheduling program according to the present invention provides a product for an initial scheduling problem regarding a manufacturing candidate given as an upper limit of the kind and number of products allowed to be manufactured in a target process given in advance at the beginning of a day. Scheduling optimization step for deriving an optimal solution for determining the type and number of the manufacturing process and the manufacturing order in the target process, and a scheduled production schedule for recording the optimal solution derived for the initial scheduling problem in the storage unit as a scheduled production schedule A recording step, a process state prediction step for predicting a process state after a certain time in the planned production schedule at a certain time after the start of operation from the planned production schedule, and an operation fluctuation expected to occur Predicted operation fluctuation recording step to record in memory as fluctuation In step state after the predetermined time predicted in step state predicting step, yet for the preparation candidates not manufactured undertaken, needs to be solved when operating fluctuations the occurrence was recorded as the estimated operational fluctuation is expected occurs A rescheduling problem prediction step for creating a rescheduling problem that occurs in response to an operation fluctuation that is expected to occur, and an optimum for determining the type and number of products and the manufacturing order in the target process for the rescheduling problem A re-scheduling problem optimal solution creation recording step for deriving a solution, and recording the derived optimal solution in a storage unit corresponding to the expected operational fluctuation as an expected re-scheduling problem optimal solution; After that, the operation fluctuation that is expected to occur and recorded as the expected operation fluctuation will actually occur. If, with the rescheduling problem optimal solution the generating recorded in the storage unit corresponding to the operation expected variations, the planned production schedule changing step of overwriting the planned production schedule to be recorded in the storage unit, the In the computer, the process state is read by the calculation unit and the process of each step is executed, and the process state prediction step, the rescheduling problem prediction step, and the rescheduling problem optimal solution creation recording step are repeatedly executed at regular intervals. It is characterized by that.

A scheduling system according to the present invention is a scheduling system that performs rescheduling in a computer having a calculation unit, a storage unit, an input unit, and an output unit, and the calculation unit is a target process given in advance at the beginning of a day Deriving the optimal solution for determining the type and number of products and the order of manufacturing in the target process for the initial scheduling problem for the manufacturing candidates given as the upper limit of the types and number of products allowed to be manufactured in A scheduling optimization unit that performs the scheduled production schedule recording unit that records the optimal solution derived for the initial scheduling problem in the storage unit as a scheduled production schedule, and at a certain point after the start of operation , A process for predicting the process state after a certain time from the scheduled production schedule. A state prediction section, the occurrence expected operating fluctuations, expected and anticipated operational fluctuation recording unit that records in the storage unit as operational fluctuations, in the process state of the after the predetermined time predicted in the step state predicting unit, still producing undertake For the manufacturing candidates that are not yet created, a rescheduling problem that needs to be solved when the expected operational fluctuations recorded as the expected operational fluctuations occur is created corresponding to the expected operational fluctuations Re-scheduling problem prediction unit, and for the re-scheduling problem , an optimal solution for determining the type and number of products and the manufacturing order in the target process is derived, and the derived optimal solution is assumed as the optimal rescheduling problem optimal solution and rescheduling problem optimal solution created recording unit that records in the storage unit in response to operation variations which the generation is expected, the constant After between said expected when operating recording the generated operation expected variations were as variation actually occurs, the rescheduling problem optimal solution the generating recorded in the storage unit corresponding to the operation expected variations A scheduled production schedule changing unit for overwriting the scheduled production schedule recorded in the storage unit, the process state prediction unit, the rescheduling problem prediction unit, the rescheduling problem optimal solution creation recording unit, It is repeatedly executed at regular intervals.

As a result, manufacturing candidates that have not yet started manufacturing based on the process state after a certain time predicted from the scheduled production schedule in which the type and number of products and the manufacturing order in the target process are determined as the optimal solution in advance. A rescheduling problem corresponding to the operational fluctuation that is expected to occur is created, optimization calculation is performed to determine the type and number of products and the manufacturing order in the target process, and the optimal solution is stored. Therefore, when actual predicted operational fluctuations occur, that is, when a rescheduling problem occurs, the stored optimal solution is overwritten on the scheduled production schedule and used as a rescheduling problem. In contrast, the optimal solution can be used in real time.
In addition, the optimum solution of the rescheduling problem when an operation fluctuation that is expected to occur is always recorded. Therefore, the optimal solution (reschedule) is maintained when operation fluctuations that are expected to occur all day occur, and the optimal solution (reschedule) is assumed when the operation fluctuations that are expected to occur are actually generated and necessary. ) Can be used.

  In addition, the scheduling method and the scheduling program according to the present invention further include a preparation time recording step of recording a preparation time necessary for start of manufacture in a storage unit for each product manufactured in the target process, and the process state prediction In the step, the certain point in time is a time that goes back the preparation time of the product from a manufacturing start time of a product that is scheduled to be manufactured next in the scheduled production schedule, and the predetermined time later is the scheduled production schedule. It may be a time that goes back the preparation time of the product from the production start time of the product that is scheduled to be manufactured next.

  Further, in the scheduling system according to the present invention, the calculation unit further includes a preparation time recording unit that records a preparation time necessary for start of manufacture in a storage unit for each product manufactured in the target process, In the state prediction unit, the certain point in time is a time that goes back the preparation time of the product from a manufacturing start time of a product that is scheduled to be manufactured next in the scheduled production schedule, and after the certain time, the scheduled production It may be a time that goes back the preparation time of the product from the production start time of the product scheduled to be manufactured next in the schedule.

  Thereby, rescheduling can be started at an appropriate time, and the target time for predicting the process state can also be appropriately taken.

Further, in the scheduling method and the scheduling program according to the present invention, the predicted operation variation recording step further records the operation variation expected to occur in the storage unit as the predicted operation variation, in combination with the accuracy, A required calculation time recording step of recording a standard calculation time for deriving an optimal solution for the rescheduling problem in the storage unit as a required calculation time, and rescheduling between the certain time and the certain time later Deriving at the re-scheduling calculation possible number derivation step for deriving the re-scheduling calculation possible number of times by calculating the re-schedulable calculation time and dividing the calculated re-scheduling calculation possible time by the necessary calculation time wherein the number of rescheduling computable number, said pre you An operation fluctuation extraction step for extracting operation fluctuations expected to occur in the order of the accuracy recorded in the operation fluctuation, and further comprising the rescheduling problem prediction step in the process state predicted in the process state prediction step, A rescheduling problem that needs to be solved when the operation fluctuation occurs may be created in response to the operation fluctuation that is expected to be extracted in the predicted operation fluctuation extraction step.

Further, in the scheduling system according to the present invention, the predicted operation variation recording unit further records the operation variation expected to occur in the storage unit as a predicted operation variation in combination with the accuracy, and the calculation unit A required calculation time recording unit for recording a standard calculation time for deriving an optimal solution for the rescheduling problem in the storage unit as a required calculation time, and between the certain time and the predetermined time later Re-schedulable calculation time deriving unit that obtains rescheduling calculation time, divides the obtained rescheduling calculation time by the necessary calculation time, and derives the rescheduling calculation possible number, and the rescheduling calculation possible number deriving unit in the number of the rescheduling calculable number derived, generated in the order of the recorded expected operating fluctuations accuracy An expected operation fluctuation extracting unit that extracts an expected operation fluctuation; and further, the occurrence of the rescheduling problem prediction unit extracted by the predicted operation fluctuation extraction unit in the process state predicted by the process state prediction unit is expected A rescheduling problem may be created that corresponds to the operational fluctuations that need to be solved when the operational fluctuations occur.

  This increases the possibility that the optimal solution of the rescheduling problem can be effectively utilized by pre-holding the optimal solution of the rescheduling problem against the operational fluctuations in which multiple occurrences are expected in the order of the probability of occurrence. .

  Here, the accuracy is an accuracy estimation rule set in advance in consideration of production management characteristics extracted from the current production schedule of the target process or the current production schedule of the process related to the target process. May be derived on the basis.

  Thus, by deriving the accuracy of operational fluctuations that are expected to occur during rescheduling from production management characteristics and accuracy estimation rules, they are expected to occur in descending order of probability of occurrence in accordance with the actual situation. Therefore, rescheduling is performed based on the extracted operation fluctuation that is expected to occur, so that there is a high possibility that the optimum solution of the rescheduling problem can be effectively utilized. Here, the process related to the target process indicates, for example, an upper process of the target process, a lower process of the target process, and a parallel process of the target process (processes flowing in parallel when there are a plurality of lines).

  The scheduling program according to the present invention can be distributed by being recorded on a removable recording medium such as a CD-ROM (Compact Disc Read Only Memory) and a DVD (Digital Versatile Disk) or a fixed recording medium such as a hard disk. It can be distributed via a communication network such as the Internet by wired or wireless telecommunication means.

  The scheduling method, the scheduling program, and the scheduling system according to the present invention create a rescheduling problem corresponding to an operation fluctuation that is expected to occur based on a process state predicted from a scheduled production schedule in advance, and perform optimization calculation. And save the optimal solution. Therefore, when the predicted operational fluctuation occurs, that is, when the rescheduling problem occurs, the stored optimal solution is used as a solution, so that the rescheduling problem is optimal in real time. There is an advantage that a solution can be used.

  The best mode for carrying out a scheduling method, a scheduling program, and a scheduling system according to the present invention will be described below with reference to a specific example with reference to the drawings.

  First, a scheduling method, a scheduling program, and a production scheduling target to which the scheduling system according to the present embodiment is applied will be described below with reference to FIG. Note that what is described below is merely an example for explanation, and does not indicate the application limit of the scheduling method, the scheduling program, and the scheduling system according to the present invention.

  As shown in FIG. 1, the process of creating production scheduling is called a target process. Here, there is one machine in the target process. Then, there is an upper process for supplying “semi-finished products” to the target process, and what is manufactured in the target process is called “product”, and the product is shipped as it is. Although there are various types of products, all types of products are manufactured from the same intermediate product, and the manufacturing time in the machine is also the same.

  In the target process, the types of products allowed to be manufactured that day and the upper limit of the number of products are given, and these are called manufacturing candidates. The problem of deciding the type and number of products to be actually manufactured from the production candidates (as a result, products not to be produced are also decided) and the order of production in those machines is also a problem in this example. The production scheduling problem shown.

  Next, based on FIG. 2, functions of the scheduling method, the scheduling program, and the scheduling system according to the present embodiment will be described. As shown in FIG. 2, the manufacturing candidate is input to the scheduling method, the scheduling program, and the scheduling system according to the present embodiment, and the result determined as the production scheduling problem (the product manufacturing order in the target process on the corresponding day). A production schedule and a production candidate determined not to be produced on the corresponding day are output.

Here, a concrete example of the constraint condition of the production scheduling problem is shown. The scheduling method, the scheduling program, and the scheduling system according to the present embodiment can be applied not only to the constraints of the production scheduling problem described below but also to the constraints of all types of production scheduling problems.
(1) Restriction condition 1 Manufacturing number upper limit value In this embodiment, the number of intermediate products supplied from the upper process varies daily with respect to the target process. Therefore, the upper limit of the number of products that can be manufactured in the target process varies from day to day.
(2) Restriction 2 Possibility of continuous production In a machine to which this embodiment is applied, there are products that can be manufactured next to a certain type of product and products that cannot be manufactured. As a specific example, the product types are 10 types from A to J, and Table 1 shows the possibility of continuous production. In the example shown in Table 1, when the row is E and the column is F, the value is described as 1, which indicates that the type F can be manufactured next to the type E. Further, when the row is F and the column is E, the value is described as 0, which indicates that the type E cannot be manufactured after the type F. In this example, as shown in Table 1, when the types are arranged in alphabetical order, the product after the ranking can be manufactured, but the preceding product cannot be manufactured from the product after the ranking. Moreover, even if the order is the product after the previous product, it indicates that the product cannot be manufactured if the order is 3 or more.
(3) Constraint 3 Delivery date Some manufacturing candidates to which this embodiment is applied must be manufactured on that day.

Next, a specific example of an evaluation index for the production scheduling problem is shown. The scheduling method, the scheduling program, and the scheduling system according to the present embodiment can be applied not only to the production scheduling problem evaluation index described below but also to any type of production scheduling problem evaluation index.
(1) Evaluation index 1 Production quantity X
Evaluation index 1 is an evaluation index in which it is important to maximize the number of products actually manufactured. The value of the production number is assumed to be X. As described above, in this embodiment, the manufacturing time of each product is the same. In addition, the upper limit of the daily operation time is assumed to be sufficiently larger than the production time × the number of production, and is not a restriction. For this reason, the maximum value of X is the production upper limit number indicated by the constraint condition 1 described above, and in this embodiment, a permutation is created from the production candidates so as to satisfy the constraint conditions 2 and 3 described above. This is a problem.
(2) Evaluation index 2 Setup change cost Y
The evaluation index 2 places importance on minimizing the setup change cost that occurs when different products are successively produced. Here, the setup change cost is shown in Table 2. Table 1 shows whether or not it can be made continuously, whereas Table 2 shows the setup change cost Y at that time. As shown in Table 2, where the value is 0 in Table 1, the value is infinite to indicate that continuous production is not possible.

The above two evaluation indexes 1 and 2 are collectively set as follows.
Minimize (-X + ωY)
Here, ω is a constant and is a weight indicating the importance of the setup change cost.

  Next, specific examples of one production scheduling problem for a day for a product manufactured in the target process and two rescheduling problems that occur due to operational fluctuations are shown. The scheduling method, scheduling program, and scheduling system according to the present embodiment are not limited to the production scheduling problem and the rescheduling problem described below, and can be applied to all types of production scheduling problems and rescheduling problems.

First, a specific example of the production scheduling problem for one day is shown. The production scheduling problem for one day is given at the beginning of the day and is called an initial scheduling problem.
<Initial scheduling problem>
○ Manufacturing candidates One by one from product A to product J ○ Manufacturer upper limit 8 pieces ○ Restriction on delivery date Items that need to be made on that day: 1 product A and 1 product J

  When such a production scheduling problem is given, an optimal solution can be derived using, for example, an optimization algorithm shown in FIG. This optimization algorithm differs depending on the production scheduling problem, and the optimization algorithm shown in FIG. 3 is merely an example, and the conventional technique is used.

One of the optimal solutions derived using the optimization algorithm in this production scheduling problem is shown below. For the initial scheduling problem, the derived optimal solution is used as the initial schedule.
<Initial schedule 1>
“A, C, D, E, F, G, I, and J are manufactured in this order, and B and H are not manufactured.”
Assume that this initial schedule 1 is adopted as the scheduled production schedule for the day. In addition, the result derived | led-out initially by the production schedule shall be called a schedule production schedule. That is, the scheduled production schedule means a schedule that is derived in the initial stage and is prepared as a basic form for a future schedule first. And things other than the result derived | led-out initially, for example, used conceptually, or what is implemented sequentially after the initial stage is called a production schedule.

  Next, a concrete example of the rescheduling problem caused by operational fluctuations is shown. Here, as an example of the operational fluctuation, an increase or decrease in the number of intermediate products supplied from the upper process can be considered. Also, if another process that uses the same intermediate product stops due to some trouble, the intermediate product that was planned to be used there will be sent to the target process and increased. In addition, although operational fluctuations such as equipment stop troubles in the target process itself and quality abnormalities can be considered, it can be dealt with in the same way as the rescheduling problem in the case where an increase or decrease in intermediate products shown below occurs.

[Example of rescheduling problem caused by operational fluctuation 1]
As an example of the operational fluctuation, for example, consider a rescheduling problem in the case where an operational fluctuation occurs in which one usable intermediate product is reduced when E is manufactured. In this case, the number of products that can be manufactured later is 3, and F, G, I, and J that have not yet been manufactured cannot be manufactured, and rescheduling is necessary.
The rescheduling problem at that time is as follows.
<Rescheduling problem 1>
○ Manufacturing candidates F, G, I, J, which have not yet started manufacturing, and B, H which were not manufactured in the initial schedule are also manufacturing candidates.
○ Upper limit of production quantity: 3 ○ Delivery date constraint In the initial scheduling problem, J that had to be created within the day has not been created yet, so it is necessary to produce this in the rescheduling problem.

The constraint condition and the evaluation index are as described above. This rescheduling problem has the same format as the initial scheduling problem, and an optimal solution can be derived using the optimization algorithm shown in FIG. An example of the result of the optimal solution derived by rescheduling is shown below. For the rescheduling problem, the derived optimal solution is rescheduled and becomes a new scheduled production schedule.
<Optimal reschedule 1>
“Create F, H, J in this order, and do not manufacture B, G, I”

[Example 2 of rescheduling problem caused by operational fluctuations]
Next, as another example of the operation fluctuation, for example, consider a rescheduling problem when an operation fluctuation occurs in which one available intermediate product increases when E is manufactured. In this case, the number of products that can be manufactured afterwards will be 5, select 5 from 6 types of products including F, G, I, J, which has not yet started manufacturing, and B, H which has not been manufactured, It becomes a problem to determine the permutation.
The rescheduling problem at that time is as follows.
<Rescheduling problem 2>
○ Manufacturing candidates F, G, I, J, which have not yet started manufacturing, and B, H which were not manufactured in the initial schedule are also manufacturing candidates.
○ Upper limit of production quantity: 5 ○ Delivery time constraint In the initial scheduling problem, J that had to be created on the same day has not been created yet, so it is necessary to produce this in the rescheduling problem.

The constraint condition and the evaluation index are as described above. This rescheduling problem has the same format as the initial scheduling problem, and an optimal solution can be derived using the optimization algorithm shown in FIG. An example of the result of the optimal solution derived by rescheduling is shown below.
<Optimal reschedule 2>
“F, G, H, I, J are created in this order, and B is not manufactured.”

  In the case described above, in order to facilitate understanding, the constraints and evaluation indexes shown in Tables 1 and 2 are simplified, and the number of products that can be manufactured is reduced to a few. Therefore, these optimum solutions can be derived in a short time by humans or optimization algorithms. However, in the production scheduling problem or the rescheduling problem in an actual production line, the number of products reaches several hundreds, and the constraints and evaluation indexes become complicated, so that the calculation time is very long. Therefore, it is difficult to derive and use the optimal solution by the rescheduling problem when the operation fluctuation occurs.

  Next, the scheduling method, scheduling program, and scheduling system according to the present embodiment will be described. In the present embodiment, it is assumed that the manufacturing start time is 8 o'clock and that one manufacturing time is 1 hour regardless of the type.

[First embodiment]
First, the scheduling system according to the first embodiment will be described with reference to FIG. FIG. 4 is a block diagram of the scheduling system according to the first embodiment. The scheduling system 1 includes an arithmetic unit 10, a storage unit 20, an input unit 30, and an output unit 40, and is mounted on a computer. Here, each part 10-40 of the scheduling system 1 shown by FIG. 1 is comprised by computers, such as a general purpose personal computer. Such a computer stores hardware such as a CPU, ROM, RAM, hard disk, and CD-ROM drive, and the hard disk includes a program (this program is a removable storage medium such as a CD-ROM or DVD). In this case, it is possible to install the software on various computers. And these parts 10-40 are constructed | assembled by combining these hardware and software.

  As shown in FIG. 4, the storage unit 20 includes an initial scheduling problem 21, an initial optimal solution 22, a scheduled production schedule 23, a predicted process state 24, an expected operation fluctuation 25, a rescheduling problem 26, And the rescheduling problem optimal solution 27.

The initial scheduling problem 21 is a production scheduling problem for one day for a product manufactured in the target process, and is given at the beginning of the day and registered in advance. As described above, the production scheduling problem is a problem in which the type and number of products to be actually manufactured are determined from the manufacturing candidates, and the manufacturing order in those machines is further determined. The initial scheduling problem 21 is input from the input unit 30 (keyboard or the like), is recorded on a removable storage medium such as a CD-ROM or DVD, is given, or is a communication network such as the Internet by wired or wireless telecommunication means. Or stored in a storage device (auxiliary storage device) such as a ROM (Read Only Memory), a RAM (Random Access Memory), or a hard disk.
In the present embodiment, the above-described <initial scheduling problem> is registered as the initial scheduling problem 21.

  The calculation unit 10 includes a scheduling optimization unit 11, a scheduled production schedule recording unit 12, a process state prediction unit 13, a predicted operation fluctuation recording unit 14, a rescheduling problem prediction unit 15, and a rescheduling problem optimal solution creation record. It comprises means 16 and scheduled production schedule changing means 17.

The scheduling optimization unit 11 is for deriving an optimal solution for the input initial scheduling problem 21. Specifically, the scheduling optimization means 11 can be realized by an optimization algorithm of an enumeration method that searches for all possible optimal solutions as shown in FIG. In addition, the calculation speed may be increased by using a branch and bound method, a simulated annealing method, or a genetic algorithm method. In addition, using an optimization algorithm as shown in FIG. 3, a method may be considered in which an upper limit is set for the calculation time, the calculation is stopped within a certain time, and the optimum solution at that time is used as the solution.
In the present embodiment, it is assumed that scheduling optimization means 11 is used that derives an optimal solution by using an optimization algorithm as shown in FIG. The scheduling optimization unit 11 derives the above-described <initial schedule 1> as an optimal solution for the above-described <initial scheduling problem>.
The optimal solution derived by the scheduling optimization unit 11 is input to the scheduled production schedule recording unit 12 as the initial optimal solution 22.

The scheduled production schedule recording means 12 is for storing the initial optimum solution 22 derived by the scheduling optimization means 11 as a scheduled production schedule 23 in the storage unit of the computer.
In the present embodiment, the derived initial optimum solution 22 <initial schedule 1> is stored as the scheduled production schedule 23 in the storage unit of the computer.

The process state predicting means 13 is for predicting a process state after a certain time from the scheduled production schedule 23 at a certain time after the start of operation.
In this embodiment, based on <Initial schedule 1> shown in FIG. 5 saved as the scheduled production schedule 23, it is possible to predict which product is created or when it is created. For example, based on FIG. 5, it can be seen that the product C is being manufactured at 9:30 and the product F is being manufactured at 12:30. Then, if the “some time after the start of operation” is, for example, 10:30 and the fixed time M = 60 minutes, the target process indicates that the product E is being manufactured, and that the product already manufactured is It can be predicted that the products scheduled to be manufactured are F, G, I, and J, and products B and H that are not scheduled to be produced in the current scheduled production schedule.
The process state predicted by the process state prediction unit 13 is input to the rescheduling problem prediction unit 15 as the predicted process state 24.

The predicted operation fluctuation recording means 14 is for recording the expected operation fluctuation in the storage unit 20 of the computer as the expected operation fluctuation 25. Operational fluctuations that are expected to occur are input from the input unit 30, are recorded on a removable storage medium such as a CD-ROM or DVD, and are given to a communication network such as the Internet by wired or wireless telecommunication means. Or is recorded in advance in the storage unit of the computer as the predicted operation fluctuation 25. A plurality of operation fluctuations that are expected to occur can be recorded.
In the present embodiment, “an intermediate product is reduced by one” is recorded in the predicted operation fluctuation 25 in advance as an operation fluctuation that is expected to occur.

The rescheduling problem prediction means 15 needs to be solved when an operation fluctuation that is expected to be recorded as the expected operation fluctuation 25 by the prediction recording means 14 occurs in the process state predicted by the process state prediction means 13. This is to create a problem corresponding to the operational fluctuations that are expected to occur.
In the present embodiment, in the process state after a certain time M = 60 minutes at 10:30 after the start of operation predicted by the process state predicting unit 13, the predicted operation change recording unit 14 records the predicted operation change 25 as “intermediate”. <Rescheduling problem 1> described above is created from the operational fluctuation that is expected to occur when one product decreases.
The created rescheduling problem is input to the rescheduling problem optimal solution creation recording means 16 as a rescheduling problem 26.

The rescheduling problem optimum solution creating / recording means 16 uses the scheduling optimizing means 11 to solve the rescheduling problem 26 input from the rescheduling problem predicting means 15, and the solution is set as the expected rescheduling problem optimum solution 27. This is to record the occurrence recorded as the expected operation fluctuation 25 corresponding to the expected operation fluctuation.
In the present embodiment, the <rescheduling problem 1> created by the rescheduling problem prediction means 15 is solved, and the optimal solution <optimal rescheduling 1> is recorded as the rescheduling problem optimal solution 27.

Here, the process state predicting means 13, the rescheduling problem predicting means 15, and the rescheduling problem optimum solution creating and recording means 16 are repeatedly executed at regular intervals, and the operation fluctuations that are expected to be recorded in the expected operation fluctuations 25 are detected. It is preferable to always record the optimal rescheduling problem solution 27 when it occurs.
In this embodiment, since the manufacturing start time is 8:00, the process state prediction means 13, the rescheduling problem prediction means 15, and the rescheduling problem optimum solution creation recording means 16 are operated every hour from 8: 1. If the optimal rescheduling problem solution 27 that is expected one hour ahead is always recorded, the optimal solution (rescheduling) in the event of an operational fluctuation that is expected to occur all day is maintained, and the expected operation is expected to occur. It becomes possible to use the optimal solution (reschedule) when fluctuations actually occur and become necessary.

The scheduled production schedule changing means 17 predicts the expected operation fluctuation 25 recorded by the rescheduling problem optimum solution creation recording means 16 when an operation fluctuation that is expected to occur as the expected operation fluctuation 25 actually occurs after a certain time has elapsed. The optimal solution for the rescheduling problem corresponding to is changed to the scheduled production schedule 23 thereafter. In other words, the scheduled production schedule changing means 17 calculates the optimal solution 27 for the expected rescheduling problem recorded in the storage unit 10 when the expected operational fluctuation recorded as the expected operational fluctuation 25 occurs after a certain period of time. The scheduled production schedule 23 is overwritten.
In the present embodiment, when an operational fluctuation “an intermediate product is reduced by one” that is expected to occur after a certain time has occurred, the <optimum reschedule 1> stored as the expected rescheduling problem optimal solution 27 is stored. The scheduled production schedule 23 is overwritten.

  The scheduled production schedule 23 recorded in the storage unit 20 is output from the output unit 40 (an output device such as a display or a printer) via the output interface. Here, the output interface is a program or hardware that performs data conversion and communication for outputting the scheduled production schedule 23 to the output device.

Next, the processing procedure of the scheduling method according to the first embodiment will be described with reference to FIG. FIG. 6 is a flowchart illustrating the processing procedure of the scheduling method according to the first embodiment.
Note that the processing of the scheduling method according to this embodiment described below can be similarly read and executed by the CPU as a scheduling program in a computer. Further, the scheduling program can be installed in storage devices of various computers by recording it in a removable storage medium such as a CD-ROM or DVD.

As shown in FIG. 6, in a computer, a communication network such as the Internet is input by an input unit, recorded on a removable storage medium such as a CD-ROM or DVD, or wired or wireless telecommunication means. An optimal solution is derived for the initial scheduling problem given through the computer, and is stored in the storage unit of the computer as a scheduled production schedule (step S1: scheduling optimization step, scheduled production schedule recording step). The initial scheduling problem is a production scheduling problem for one day for a product manufactured in the target process, and is given at the beginning of the day and registered in advance.
The details of this step are the same as those described in the scheduling optimization unit 11 and the scheduled production schedule recording unit 12 of the scheduling system 1 described above, and a description thereof will be omitted.

  Then, it is determined whether or not rescheduling is started, that is, whether or not a certain point in time after the start of operation has been reached (step S2). If it is not at a certain time after the start of operation (step S2: N), the process returns to step S2. On the other hand, if it is at a certain time after the start of operation (step S2: Y), the process proceeds to step S3.

At a certain time after the start of operation (step S2: Y), the process state after a certain time is predicted from the scheduled production schedule stored in step S1 (step S3: process state prediction step).
The details of this step are the same as those described in the process state predicting means 13 of the scheduling system 1 described above, and a description thereof is omitted.

  Then, it is determined whether or not all production has been completed (step S4). If all production is not completed (step S4: N), the process proceeds to step S5. On the other hand, if all the production is finished (step S4: Y), the process is finished.

If the entire production has not been completed (step S4: N), a rescheduling problem that needs to be solved when an operation fluctuation that is expected to occur in the process state predicted in step S3 occurs is created (step S5: Forecast operation fluctuation recording step, rescheduling problem forecast step). Note that the operation fluctuation that is expected to occur is recorded in advance as the expected operation fluctuation.
The details of this step are the same as those described in the predicted operation fluctuation recording means 14 and the rescheduling problem prediction means 15 of the scheduling system 1 described above, and a description thereof will be omitted.

Then, the rescheduling problem created in step S5 is solved, and the solution is recorded as the expected rescheduling problem optimum solution (step S6: rescheduling problem optimum solution creation recording step).
The details of this step are the same as those described in the re-scheduling problem optimum solution creating / recording means 16 of the scheduling system 1 described above, and a description thereof will be omitted.

  Next, it is determined whether or not a certain time has passed (step S7). If the predetermined time has not elapsed (step S7: N), the process returns to step S7. On the other hand, if the predetermined time has elapsed (step S7: Y), the process proceeds to step S8.

  If the predetermined time has elapsed (step S7: Y), it is determined whether or not an operation fluctuation that has been predicted to be recorded in advance as an expected operation fluctuation in step S5 has occurred (step S8). That is, after a predetermined time has elapsed, it is determined in step S5 whether or not an operation fluctuation that is actually expected to occur as an expected operation fluctuation has occurred. When the operation fluctuation that is expected to occur does not occur (step S8: N), the process returns to step S2. On the other hand, when an operation fluctuation that is expected to occur occurs (step S8: Y), the process proceeds to step S9.

When an operation fluctuation that is expected to occur occurs (step S8: Y), the optimal solution for the rescheduling problem recorded in step S6 is changed to the subsequent scheduled production schedule (step S9: scheduled production schedule changing step). . That is, when an operation fluctuation that is expected to occur occurs, the optimal solution for the expected rescheduling problem recorded in the storage unit is overwritten on the scheduled production schedule recorded in step S1.
The details of this step are the same as the contents described in the scheduled production schedule changing means 17 of the scheduling system 1 described above, and the description thereof is omitted.

  As described above, according to the scheduling system, the scheduling method, and the scheduling program of the first embodiment, even when an operation fluctuation that is expected to occur occurs, an operation fluctuation that is expected to occur always occurs. The optimal solution (reschedule) can be used, and benefits such as improved productivity, improved delivery rate, and cost reduction can be expected.

[Second Embodiment]
Next, a scheduling system according to the second embodiment will be described with reference to FIG. FIG. 8 is a block diagram of a scheduling system according to the second embodiment. In addition, about the site | part similar to 1st embodiment, the same code | symbol is attached | subjected and the description is abbreviate | omitted.

  As shown in FIG. 8, in the scheduling system 1 according to the second embodiment, in addition to the scheduling system 1 according to the first embodiment, the calculation unit 10 includes a preparation time recording unit 18, and the storage unit 20 includes Preparation time 28 is provided.

The preparation time recording means 18 is for recording in the storage unit 20 the preparation time necessary for the start of production for each product manufactured in the target process as the preparation time 28. Here, the preparation time means a time for arranging materials or tools for work or a person, for example, although the target process is not actually used. The preparation time 28 recorded in the storage unit 20 is input to the process state predicting means 13 as shown in FIG. FIG. 7 is a diagram illustrating the rescheduling execution time. As shown in FIG. 7, in order to start production of the product C, it is necessary to start its preparation 30 minutes before, and 20 minutes are necessary to start production of the product D. .

The process state prediction means 13 is for predicting the process state after a certain time from the scheduled production schedule 23 at a certain time after the start of operation, as in the first embodiment. Here, “a certain point in time” after the start of operation is the time when rescheduling is started, that is, the time when the preparation time 28 of the product is traced back from the production start time of the product scheduled to be manufactured next in the scheduled production schedule 23. In addition, “after a certain time” for predicting the process state is the time when the next rescheduling is started, that is, from the manufacturing start time of the next product scheduled to be manufactured in the scheduled production schedule 23. It is the time that goes back the product preparation time 28.
In this embodiment, as shown in FIG. 7, the production of the product C in the target process starts at 9:00, but the preparation is started at 8:30, so the scheduled production schedule is already changed. It is not possible. Accordingly, the manufacture of the product C is confirmed at 8:30, and the scheduled production schedule can be changed after the next product D. Since the preparation time of the product D is 20 minutes, the manufacture of the product D is confirmed at 9:40. Therefore, even if rescheduling performed at a later time, for example, 9:50, is prepared in anticipation of operational fluctuations, it becomes invalid. In other words, when the production of the product C is confirmed at 8:30, rescheduling is performed, and at that time, 9:40 can be said to be appropriate as the time for predicting the occurrence of operation fluctuation. To generalize this, it is preferable that the rescheduling predicts immediately after the next product to be manufactured in the scheduled production schedule is determined until the next product to be manufactured is determined. In other words, rescheduling based on the prediction is started at a time that goes back the preparation time of the product to be manufactured next in the scheduled production schedule, and the predicted operation fluctuation time is the next in the scheduled production schedule. It can be seen that it is better to set the preparation time of the product retroactively from the production start time of the product to be manufactured.

Next, the processing procedure of the scheduling method according to the second embodiment will be described with reference to FIG. FIG. 9 is a flowchart illustrating the processing procedure of the scheduling method according to the second embodiment. In addition, about the process similar to the scheduling method which concerns on 1st embodiment, the same code | symbol is attached | subjected and description of the process content is abbreviate | omitted.
Note that the processing of the scheduling method according to this embodiment described below can be similarly read and executed by the CPU as a scheduling program in a computer. Further, the scheduling program can be installed in storage devices of various computers by recording it in a removable storage medium such as a CD-ROM or DVD.

In the second embodiment, step S2 and step S3 of the first embodiment are different. That is, in the second embodiment, after step S1, the preparation time required for the start of manufacture is recorded for each product manufactured in the target process (step S11: preparation time recording step).
The details of this step are the same as those described in the preparation time recording means 18 of the scheduling system 1 described above, and a description thereof will be omitted.

Whether or not the rescheduling start time has been reached, that is, whether or not a certain point in time after the start of operation has reached a time Ti that goes back the preparation time of the product from the manufacturing start time of the next manufactured product in the scheduled production schedule Judgment is made (step S12). If the rescheduling start time Ti is reached (step S12: Y), the process state at time T (i + 1) that goes back the preparation time of the product to be manufactured next in the scheduled production schedule. Is predicted from the scheduled production schedule stored in step S1 (step S13: process state prediction step).
The details of this step are the same as those described in the process state predicting means 13 of the scheduling system 1 described above, and a description thereof is omitted.

  As described above, according to the scheduling system, the scheduling method, and the scheduling program of the second embodiment, rescheduling can be started at an appropriate time, and the target time for predicting the process state can be appropriately taken. Can do.

[Third embodiment]
Next, a scheduling system according to the third embodiment will be described with reference to FIG. FIG. 10 is a block diagram of a scheduling system according to the third embodiment. In addition, about the site | part similar to 1st embodiment and 2nd embodiment, the same code | symbol is attached | subjected and the description is abbreviate | omitted.

  As shown in FIG. 10, in the scheduling system 1 according to the third embodiment, in addition to the scheduling system 1 according to the second embodiment, the calculation unit 10 obtains the necessary calculation time recording means 19 and the number of rescheduling calculation possible times. Means 111 and predicted operation fluctuation extraction means 112 are provided, and the storage unit 20 is provided with a necessary calculation time 29.

The necessary calculation time recording means 19 measures the calculation time when rescheduling and records the standard time as the required calculation time 29 in the storage unit 20. Here, the calculation time for performing rescheduling indicates the calculation time for solving the rescheduling problem 26 input from the rescheduling problem predicting means 15 using the scheduling optimization means 11. The necessary calculation time 29 is input to the rescheduling calculation possible time extraction unit 111.
In this embodiment, it is assumed that it is 15 minutes, for example.

The rescheduling calculation possible number deriving means 111 is from the time when rescheduling is started to the time when rescheduling is started next, that is, from a certain time after the start of operation until a certain time after predicting the process state. That is, the manufacture of the next product scheduled to be manufactured in the scheduled production schedule 23 is started from the time after the preparation time 28 of the product that is scheduled to be manufactured next in the scheduled production schedule 23. The re-schedulable calculation time is calculated from the time until the time when the product preparation time 28 is traced back, and the calculated re-scheduling calculation time is divided by the required calculation time 29 to derive the re-schedulable calculation time. Is to do.
In this embodiment, for example, when FIG. 7 is used as an example, the rescheduling calculation time needs to start at 8:30 for the first rescheduling and end by 9:40, for example. Therefore, the first rescheduling calculation possible time is 70 minutes. Similarly, the second rescheduling calculation time is 50 minutes, the third rescheduling calculation time is 40 minutes, and the fourth rescheduling calculation time is 1 hour and 40 minutes. The first reschedulable calculation count is 70 minutes / 15 = 4 plus 10 minutes, the second rescheduling calculation count is 50 minutes / 15 = 3 plus 5 minutes, and the third rescheduling calculation count is 40. Minute ÷ 15 = 2 plus 10 minutes, 4th rescheduling calculation possible time is 1 hour 40 minutes ÷ 15 = 6 plus 10 minutes, and 4 times, 3 times, 2 times and 6 times of rescheduling are possible respectively Become.

The expected operation fluctuation recording means 14 is for recording the operation fluctuation expected to occur in the storage unit 20 of the computer as the expected operation fluctuation 25 in combination with the accuracy. Operation fluctuations expected to occur and their accuracy are input from the input unit 30, recorded in a removable storage medium such as a CD-ROM or DVD, or given by wired or wireless telecommunication means such as the Internet. For example, the predicted operation fluctuation 25 is recorded in advance in the storage unit of the computer.
Table 3 shows examples of operation fluctuations and their accuracy that are expected to occur in this embodiment.

The predicted operation fluctuation extracting unit 112 extracts operation fluctuations that are expected to occur in the order of the accuracy recorded in the predicted operation fluctuation 25 by the number of rescheduling calculation possible times derived by the rescheduling calculation possible number deriving unit 111. That is, on the basis of the expected operation fluctuation 25, the operation fluctuation that is expected to be generated with higher accuracy by the number of rescheduling calculation possible times derived by the rescheduling calculation possible number deriving means 111 is extracted. . The extracted operation fluctuations that are expected to occur are input to the rescheduling problem prediction means 15.
In the present embodiment, for example, when FIG. 7 is used as an example, in the case of the first rescheduling, the number of rescheduling calculations is four, so four types of calculations can be performed, and all the operational fluctuations in Table 3 are extracted. To do. In the case of the second rescheduling, since the number of rescheduling calculations is three, three types of calculations can be performed, and operation fluctuations A to C in Table 3 are extracted. In the case of the third rescheduling, since the number of rescheduling calculations is two, two types of calculations can be performed, and operation fluctuations A and B in Table 3 are extracted. In the case of the fourth rescheduling, since the number of rescheduling calculations is six, six types of calculations can be performed, and all the operational fluctuations in Table 3 are extracted.

  The rescheduling problem predicting means 15 corresponds to the operation fluctuation that is expected to occur in the process state predicted by the process state prediction means 13 and is predicted to occur in response to the operation fluctuation that is expected to occur. This is to create a rescheduling problem that needs to be solved if it occurs.

Next, a processing procedure of the scheduling method according to the third embodiment will be described with reference to FIG. FIG. 11 is a flowchart illustrating the processing procedure of the scheduling method according to the third embodiment. In addition, about the process similar to the scheduling method which concerns on 2nd embodiment, the same code | symbol is attached | subjected and description of the process content is abbreviate | omitted.
Note that the processing of the scheduling method according to this embodiment described below can be similarly read and executed by the CPU as a scheduling program in a computer. Further, the scheduling program can be installed in storage devices of various computers by recording it in a removable storage medium such as a CD-ROM or DVD.

In the third embodiment, steps S14 to S16 are added to the second embodiment. That is, in the third embodiment, before rescheduling, the calculation time for rescheduling is measured, and the standard time is recorded as the required calculation time (step S14: required calculation time recording step). ).
Note that the details of this step are the same as those described in the necessary calculation time recording means 19 of the scheduling system 1 described above, and a description thereof will be omitted.

If the total production has not been completed (step S4: N), the number of rescheduling calculations is derived (step S15: rescheduling calculation number deriving step).
Note that the details of this step are the same as the description of the rescheduling calculation possible number deriving means 111 of the scheduling system 1 described above, and a description thereof will be omitted.

Next, based on the possible number of rescheduling calculations derived in step S15, an operation variation that creates a rescheduling problem is extracted from the operation variation that is predicted to be recorded in pairs with the accuracy (step S16: prediction). Operation fluctuation extraction step).
The details of this step are the same as the contents described in the re-estimated operation fluctuation extracting means 112 of the scheduling system 1 described above, and the description thereof is omitted.

Then, in the process state predicted in step S13 , a rescheduling problem that needs to be solved when the operation fluctuation extracted in step S16 occurs is created (step S5: rescheduling problem prediction step).

  As described above, according to the scheduling system, the scheduling method, and the scheduling program of the third embodiment, the optimal solution of the rescheduling problem is preliminarily obtained with respect to operation fluctuations in which a plurality of occurrences are expected in the descending order of occurrence. By holding, there is a high possibility that the optimal solution of the rescheduling problem can be used effectively.

The accuracy described above is based on the accuracy estimation rule set in advance in consideration of the production management characteristics extracted from the current production schedule of the target process or the current production schedule of the process related to the target process. Preferably it is derived. Here, the process related to the target process indicates, for example, an upper process of the target process, a lower process of the target process, and a parallel process of the target process (processes flowing in parallel when there are a plurality of lines). The accuracy estimation rule is registered in advance in the storage unit 20 with the characteristics of production management as the condition part and the accuracy of the operation fluctuation expected to occur in the descriptive part, and is referred to by the expected operation fluctuation recording means 14.
The accuracy estimation rule is, for example, assuming that if the type of intermediate product currently produced by the upper process is different, the probability that a defective product will occur in the upper process or the probability that a surplus product will occur will vary. Therefore, the type of intermediate products currently produced is a feature of production management. Table 4 shows specific examples of production management characteristics.

  It is assumed that the accuracy of the operation fluctuations A to C shown in Table 3 can be determined as shown in Table 5 in correspondence with the accuracy estimation rule with the production management characteristics shown in Table 4 as conditions.

  In this way, by deriving the accuracy of operational fluctuations that are expected to occur during rescheduling from production management characteristics and accuracy estimation rules, the occurrences are predicted in the order of the most likely occurrences in accordance with the actual situation. Therefore, the rescheduling is performed based on the extracted operation fluctuation that is expected to occur, so that there is a high possibility that the optimum solution of the rescheduling problem can be effectively utilized.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and various modifications can be made as long as they are described in the claims.

  In the scheduling method according to the third embodiment described above, as shown in FIG. 11, the process is performed in the order of the scheduling optimization step, the scheduled production schedule recording step S1, the preparation time recording step S11, and the necessary calculation time recording step S14. However, the order is not limited, and any order is acceptable.

It is a figure which shows the outline | summary of a production scheduling object. It is a figure which shows the function of the scheduling method and scheduling program which concern on this embodiment, and a scheduling system. It is the flowchart explaining the procedure of the process of the optimization algorithm of a prior art. It is a block diagram of the scheduling system which concerns on 1st embodiment. It is a figure which shows a specific example of a production schedule. It is the flowchart explaining the procedure of the process of the scheduling method which concerns on 1st embodiment. It is a figure shown about the implementation time of rescheduling. It is a block diagram of the scheduling system which concerns on 2nd embodiment. It is the flowchart explaining the procedure of the process of the scheduling method which concerns on 2nd embodiment. It is a block diagram of the scheduling system which concerns on 3rd embodiment. It is the flowchart explaining the procedure of the process of the scheduling method which concerns on 3rd embodiment.

DESCRIPTION OF SYMBOLS 1 Scheduling system 10 Operation part 11 Scheduling optimization means 12 Schedule production schedule recording means 13 Process state prediction means 14 Predictive operation fluctuation recording means 15 Rescheduling problem prediction means 16 Rescheduling problem optimal solution creation recording means 17 Schedule production schedule change means 18 Preparation time recording means 19 Necessary calculation time recording means 20 Storage section 21 Initial scheduling problem 22 Initial optimum solution 23 Scheduled production schedule 24 Predicted process state 25 Expected operation fluctuation 26 Rescheduling problem 27 Rescheduling problem optimum solution 28 Preparation time 29 Necessary Calculation time 30 Input unit 40 Output unit 111 Re-scheduling calculation possible number deriving unit 112 Expected operation fluctuation extracting unit S1 Scheduling optimization step, scheduled production schedule recording unit Step S3 Process state prediction step S5 Predicted operation fluctuation recording step, rescheduling problem prediction step S6 Rescheduling problem optimal solution creation recording step S9 Scheduled production schedule change step S11 Preparation time recording step S13 Process state prediction step S14 Required calculation time recording step S15 Re-scheduling calculation possible number derivation step S16 Expected operation fluctuation extraction step

Claims (12)

In a computer having an arithmetic unit, a storage unit, an input unit, and an output unit,
The processing executed by the computing unit is
For the initial scheduling problem about the production candidate given as the upper limit of the type and number of products allowed to be manufactured in the target process given in advance at the beginning of the day, the type and number of products and the target process A scheduling optimization step for deriving an optimal solution for determining the manufacturing order of
A scheduled production schedule recording step of recording an optimal solution derived for the initial scheduling problem in a storage unit as a scheduled production schedule;
At a certain time after the start of operation, a process state prediction step for predicting a process state after a certain time in the scheduled production schedule from the scheduled production schedule;
An expected operation fluctuation recording step for recording the operation fluctuation expected to occur in the storage unit as an expected operation fluctuation;
Necessary to solve when there is an operation fluctuation that is expected to be recorded as the expected operation fluctuation for the production candidate that has not yet started production in the process state after the predetermined time predicted in the process state prediction step. A rescheduling problem prediction step for creating a rescheduling problem that occurs in response to an operational fluctuation that is expected to occur;
For the rescheduling problem, an optimal solution that determines the type and number of products and the manufacturing order in the target process is derived, and the operation that is expected to occur as the optimal solution for the rescheduling problem that is predicted Rescheduling problem optimal solution creation recording step to record in the storage unit corresponding to the fluctuation,
The rescheduling problem corresponding to the expected operational fluctuation recorded in the storage unit when the expected operational fluctuation recorded as the predicted operational fluctuation has actually occurred after the predetermined time has elapsed. A scheduled production schedule changing step of overwriting the optimal solution on the scheduled production schedule recorded in the storage unit;
The scheduling method is characterized in that the process state prediction step, the rescheduling problem prediction step, and the rescheduling problem optimal solution creation recording step are repeatedly executed at regular intervals. For each product manufactured in the target process, further comprising a preparation time recording step of recording a preparation time necessary for the start of manufacture in a storage unit,
In the process state prediction step,
The certain point in time is a time that goes back the preparation time of the product from the production start time of the product scheduled to be manufactured next in the scheduled production schedule,
2. The scheduling method according to claim 1, wherein the time after the predetermined time is a time retroactive to a preparation time of the product from a manufacturing start time of a product scheduled to be manufactured next in the scheduled production schedule. . The predicted operation fluctuation recording step further records the operation fluctuation that is expected to occur in the storage unit as the expected operation fluctuation, paired with its accuracy,
A required calculation time recording step of recording a standard calculation time for deriving an optimal solution for the rescheduling problem in the storage unit as a required calculation time;
Rescheduling calculation time is obtained by calculating a rescheduling calculation time from the certain time point to a certain time later, and dividing the calculated rescheduling calculation time by the necessary calculation time to derive the number of rescheduling calculation times The number derivation step;
Expected operation fluctuation extraction step for extracting operation fluctuations that are expected to occur in the order of the accuracy recorded in the expected operation fluctuations by the number of rescheduling calculation possibility times derived in the rescheduling calculation possibility number derivation step;
Further comprising
The rescheduling problem prediction step is solved when the operation fluctuation occurs in the process state predicted in the process state prediction step, corresponding to the operation fluctuation expected to be generated in the predicted operation fluctuation extraction step. 3. A scheduling method according to claim 1 or 2, wherein a rescheduling problem that arises is created.   The accuracy is based on an accuracy estimation rule set in advance in consideration of production management characteristics extracted from the current scheduled production schedule of the target process or the current production schedule of the process related to the target process. The scheduling method according to claim 3, wherein the scheduling method is derived. For the initial scheduling problem about the production candidate given as the upper limit of the type and number of products allowed to be manufactured in the target process given in advance at the beginning of the day, the type and number of products and the target process A scheduling optimization step for deriving an optimal solution for determining the manufacturing order of
A scheduled production schedule recording step of recording an optimal solution derived for the initial scheduling problem in a storage unit as a scheduled production schedule;
At a certain time after the start of operation, a process state prediction step for predicting a process state after a certain time in the scheduled production schedule from the scheduled production schedule;
An expected operation fluctuation recording step for recording the operation fluctuation expected to occur in the storage unit as an expected operation fluctuation;
Necessary to solve when there is an operation fluctuation that is expected to be recorded as the expected operation fluctuation for the production candidate that has not yet started production in the process state after the predetermined time predicted in the process state prediction step. A rescheduling problem prediction step for creating a rescheduling problem that occurs in response to an operational fluctuation that is expected to occur;
For the rescheduling problem, an optimal solution that determines the type and number of products and the manufacturing order in the target process is derived, and the operation that is expected to occur as the optimal solution for the rescheduling problem that is predicted Rescheduling problem optimal solution creation recording step to record in the storage unit corresponding to the fluctuation,
The rescheduling problem corresponding to the expected operation fluctuation recorded in the storage unit when the expected operation fluctuation recorded as the expected operation fluctuation after the predetermined time has actually occurred A scheduled production schedule changing step of overwriting the optimal solution on the scheduled production schedule recorded in the storage unit;
The process state prediction step, the rescheduling problem prediction step, and the rescheduling problem optimal solution creation recording step are repeatedly executed at regular intervals. A scheduling program. For each product manufactured in the target process, further comprising a preparation time recording step of recording a preparation time necessary for the start of manufacture in a storage unit,
In the process state prediction step,
The certain point in time is a time that goes back the preparation time of the product from the production start time of the product scheduled to be manufactured next in the scheduled production schedule,
The scheduling program according to claim 5, wherein the predetermined time is a time that is a time that is retroactive from a manufacturing start time of a product scheduled to be manufactured next in the scheduled production schedule. . The predicted operation fluctuation recording step further records the operation fluctuation that is expected to occur in the storage unit as the expected operation fluctuation, paired with its accuracy,
A required calculation time recording step of recording a standard calculation time for deriving an optimal solution for the rescheduling problem in the storage unit as a required calculation time;
Rescheduling calculation time is obtained by calculating a rescheduling calculation time from the certain time point to a certain time later, and dividing the calculated rescheduling calculation time by the necessary calculation time to derive the number of rescheduling calculation times The number derivation step;
Expected operation fluctuation extraction step for extracting operation fluctuations that are expected to occur in the order of the accuracy recorded in the expected operation fluctuations by the number of rescheduling calculation possibility times derived in the rescheduling calculation possibility number derivation step;
Further comprising
The rescheduling problem prediction step is solved when the operation fluctuation occurs in the process state predicted in the process state prediction step, corresponding to the operation fluctuation expected to be generated in the predicted operation fluctuation extraction step. The scheduling program according to claim 5 or 6, wherein a rescheduling problem that arises is created.   The accuracy is based on an accuracy estimation rule set in advance in consideration of production management characteristics extracted from the current scheduled production schedule of the target process or the current production schedule of the process related to the target process. The scheduling program according to claim 7, wherein the scheduling program is derived. In a computer having a calculation unit, a storage unit, an input unit, and an output unit, a scheduling system for performing rescheduling,
The computing unit is
For the initial scheduling problem about the production candidate given as the upper limit of the type and number of products allowed to be manufactured in the target process given in advance at the beginning of the day, the type and number of products and the target process A scheduling optimization unit for deriving an optimal solution for determining the manufacturing order of
An optimal solution derived for the initial scheduling problem, a scheduled production schedule recording unit for recording in a storage unit as a scheduled production schedule;
At a certain time after the start of operation, a process state prediction unit that predicts a process state after a certain time in the planned production schedule from the planned production schedule,
An expected operation fluctuation recording unit that records the expected operation fluctuation in the storage unit as an expected operation fluctuation;
Necessary to solve when there is an operation fluctuation that is expected to occur and recorded as the expected operation fluctuation for the production candidate that has not yet started production in the process state after the predetermined time predicted by the process state prediction unit. A rescheduling problem prediction unit that creates a rescheduling problem that occurs in response to an operation fluctuation that is expected to occur,
For the rescheduling problem, an optimal solution that determines the type and number of products and the manufacturing order in the target process is derived, and the operation that is expected to occur as the optimal solution for the rescheduling problem that is predicted Rescheduling problem optimal solution creation recording unit that records in the storage unit corresponding to the fluctuation,
The rescheduling problem corresponding to the expected operation fluctuation recorded in the storage unit when the expected operation fluctuation recorded as the expected operation fluctuation after the predetermined time has actually occurred A planned production schedule changing unit for overwriting the optimal solution on the planned production schedule recorded in the storage unit;
The scheduling system is characterized in that the process state prediction unit, the rescheduling problem prediction unit, and the rescheduling problem optimal solution creation recording unit are repeatedly executed at regular intervals. The computing unit is
For each product manufactured in the target process, further comprising a preparation time recording unit that records a preparation time required for the start of manufacture in a storage unit,
In the process state prediction unit,
The certain point in time is a time that goes back the preparation time of the product from the production start time of the product scheduled to be manufactured next in the scheduled production schedule,
The scheduling system according to claim 9, wherein the predetermined time is a time that is a retrospective preparation time of a product that is scheduled to be manufactured next in the scheduled production schedule. . The predicted operation fluctuation recording unit further records the operation fluctuation expected to occur in the storage unit as the predicted operation fluctuation, in combination with the accuracy,
The computing unit is
A required calculation time recording unit that records a standard calculation time for deriving an optimal solution to the rescheduling problem as a required calculation time in a storage unit;
Rescheduling calculation time is obtained by calculating a rescheduling calculation time from the certain time point to a certain time later, and dividing the calculated rescheduling calculation time by the necessary calculation time to derive the number of rescheduling calculation times A number derivation unit;
An expected operation fluctuation extraction unit that extracts the operation fluctuations that are expected to occur in the order of the accuracy recorded in the predicted operation fluctuations, as many times as the number of rescheduling calculation times derived by the rescheduling calculation possible number derivation unit;
Further comprising
The rescheduling problem prediction unit solves the case where the operation variation occurs in the process state predicted by the process state prediction unit in response to the operation variation expected to be generated by the predicted operation variation extraction unit. 11. A scheduling system according to claim 9 or 10, characterized by creating a rescheduling problem that arises. The accuracy is based on an accuracy estimation rule set in advance in consideration of production management characteristics extracted from the current scheduled production schedule of the target process or the current production schedule of the process related to the target process. The scheduling system according to claim 11, wherein the scheduling system is derived.