JP2816022B2 - Switching management method in scheduling - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION 【Overview】

A scheduling method and apparatus for arranging tasks to be scheduled in a time-series manner, aiming at uniformly managing switching occurring between tasks and improving the efficiency of the scheduling processing apparatus. , A work switching time table that stores data about the switching time from one work to another, a resource switching time table that stores data about the switching time from one resource to another, and And work standard information indicating the type of work, and when designating the placement of one work, an identification flag for the type of work is set, and the work is placed immediately before the work in chronological order. The switching time from the other work to the relevant work, and the switching time from the relevant work to the immediately succeeding work are changed according to the identification flag. When the sum of the time required for the work and the switching time is larger than the time between the immediately preceding work and the immediately succeeding work, the work is determined by referring to either the time table or the resource switching time table. Are arranged so as not to be arranged.

TECHNICAL FIELD OF THE INVENTION

The present invention relates to a scheduling method and apparatus for arranging tasks to be scheduled in time series. With the development of artificial intelligence in recent years, there is an increasing demand for solving various scheduling problems by an expert system. In response to this, various expert system construction tools are provided. For example, in a production plan at a factory, it is possible to efficiently manage the switching that occurs between works and perform a scheduling process. No building tools have been provided yet.

[Prior art]

Generally, when scheduling a production plan,
For each machine or facility operating in the factory, a time-series table is created by chronologically arranging scheduling targets such as work, work, personnel, or products (hereinafter, referred to as “work”). Is done. When arranging works in chronological order, it is often necessary to switch (or changeover) the time between works, so when creating a chronological table, the switching is performed. Time needs to be considered. In a conventional scheduling device, information on switching time is stored in a task (scheduling target). When arranging each task, a user (operator) performs scheduling while confirming whether switching has occurred. I was going.

[Problems to be solved by the invention]

Therefore, every time the work is arranged, it is necessary to check the occurrence of the switching, and if the switching occurs, it is necessary to determine the switching time and determine whether or not the arrangement is possible, and the arrangement of the work is very troublesome. There was a problem. In addition, when deleting a work due to a change in the schedule, it is necessary to consider whether or not the switching has occurred and the switching time. As described above, since the switching management is not uniform and complicated, there is a problem in improving the efficiency of the scheduling process. The present invention has been made in view of the above-described problem, and has as its object to unify management of switching that occurs between works and improve the efficiency of scheduling processing.

[Means for Solving the Problems]

The method according to the first aspect of the present invention is a scheduling processing method for arranging work WRKs to be scheduled in a time-series manner as shown in FIGS. A work switching time table 20 storing data on a switching time TCH from a certain work WRK to another work WRK, and a certain resource RES to another resource R
Resource switching time table 21 storing data on switching time TCH to ES, and resource R required for performing work WRK
Work standard information 22 indicating ES is provided and one work
When instructing the placement of the WRK, set the identification flag FG for the type of the work WRK, and switch from the other work WRK arranged immediately before the work WRK to the work WRK in chronological order. Time TCH and the work WRK immediately after the work WRK
The switching time TCH is determined by referring to either the work switching time table 20 or the resource switching time table 21 according to the identification flag FG, and the sum of the time required for the work WRK and the switching time TCH is obtained. , Last working WRK
This is a processing method for disabling the work WRK when it is longer than the time between the work WRK and the work WRK immediately after. An apparatus according to the second aspect of the present invention is a scheduling apparatus 1 for arranging work WRKs to be scheduled in time series, and stores data on a switching time TCH from one work WRK to another work WRK. Work switching time table 20, a resource switching time table 21 storing data on a switching time TCH from one resource RES to another resource RES, and work standard information 22 indicating the resource RES required for performing the work WRK. When the arrangement of the work WRK is instructed, referring to either the work switching time table 20 or the resource switching time table 21 according to the identification flag FG for the type of the work WRK, Switching time TCH from another work WRK arranged immediately before in time series to the work WRK, and / or the work WRK
From the switching time calculation means 31 to determine the switching time TCH to the immediately following work WRK, the time required for the work WRK, the switching time TCH, and the time between the immediately preceding work WRK and the immediately succeeding work WRK, A work arrangement management unit 32 for determining whether the work WRK can be arranged. When the arrangement of the work WRK is instructed, the switching time TCH according to the arrangement of the work WRK is obtained, and the obtained switching time TCH, the time required for the work WRK, the time between the immediately preceding work WRK and the immediately succeeding work WRK. It is determined whether or not the arrangement is possible according to the conditions. In determining the switching time TCH, the work switching time table 20 or the resource switching time table 21 is used.
Which of these is used is determined by the contents of the identification flag FG. When the resource switching time table 21 is used, the work standard information 22 is used as needed to obtain a resource RES required for the work WRK.

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of a knowledge base system 1 according to the present invention. The knowledge base system 1 of the present embodiment is an expert system for performing a scheduling process in a manufacturing process. The knowledge base system 1 includes a knowledge base 11, an inference engine 12, and a working memory 13. The knowledge base 11 is provided with a work switching time table 20, a resource switching time table 21, and work standard information 22. FIG. 2 is a diagram showing an example of the work switching time table 20. The work switching time table 20 is a table that stores data on a switching time TCH from one work WRK to another work WRK for a plurality of works WRKs. In the work switching time table 20, the work name is "A".
The switching time TCH between them for the work WRKs of ~~ "D" is shown. For example, switching time 1 from work “A” to work “B”
It takes time, and vice versa. Note that it is not possible to switch from work “D” to work “C”. In this specification, the work switching time table 20 shown in FIG. 2 may be referred to as a production type switching time matrix. FIG. 3 shows a resource switching time table 21a for the model machine.
FIG. 4 is a diagram showing an example of the resource switching time table 21b for the cutter. These resource switching time tables 21a and 21b are tables that store data on the switching time TCH from one resource RES to another resource RES for a plurality of resources RES. The resource switching time table 21a in FIG. 3 shows the switching time TCH between the model machines “Χ” to “Z”, which are the resource RES, between them. The resource switching time table 21b in FIG.
And "Q", the switching time TCH between them is shown. In this specification, these resource switching time tables 21a and 21b may be referred to as a resource type switching time matrix. FIG. 5 is a diagram showing an example of the work standard information 22. The work specification information 22 contains the resources RE required for the working WRK implementation.
5 shows S and its performance capability MAB. The work standard information 22 indicates the types of molds and cutters required as the resources RES for each of the works “A” to “C”, and the production capacity MAB when they are used. Returning to FIG. 1, the inference engine 12 is provided with a schedule creation processing unit 23 and a switching management unit 24. The schedule creation processing unit 23 performs generation, evaluation, selection, and modification check of the schedule draft SDR. The switching management unit 24 manages the switching CHG that occurs between the work WRK and the work WRK, so that the switching time calculation unit 31, the work placement management unit 32, the placement unit 33, and the identification flag
FG is provided. When the allocation of the work WRK is instructed from the schedule creation processing unit 23, the switching time calculation unit 31 determines which of the work switching time table 20 and the resource switching time table 21 according to the identification flag FG for the type of the work WRK. The switching time TCH from another work WRK arranged immediately before the work WRK in time series to the work WRK,
And the switching time TCH from the work WRK to the immediately following work WRK
Is calculated to obtain. The work arrangement management unit 32 compares the total of the switching times TCH calculated by the switching time calculation unit 31 with the time between the immediately preceding work WRK and the immediately following work WRK, and determines whether the work WRK can be arranged. Determine whether or not. The processing operation of the inference engine 12 will be described later in detail. The work memory 13 is provided with a time series table 25. 6 and 7 are diagrams showing examples of the time series tables 25a and 25b. In the time series tables 25a and 25b, the horizontal axis indicates the real time (time), and the work WRK selected as the schedule plan is arranged on the real time. At the same time, switching between work WRK and work WRK
Occurs, it is arranged in consideration of the switching time TCH. For example, in the time series table 25a of FIG. 6, the operation "C" is executed for 3 hours from 1 to 4 o'clock, and then the switching CHG1 for switching to the operation "A" is executed with the switching time TCH of 2 hours. However, a schedule is shown in which work "A" is executed for 4 hours from 12:00 to 16:00. However, the time series tables 25a and 25b shown in FIGS. 6 and 7 are shown for convenience on the image. Actually, for example, the time series data for each work WRK is successively indicated by pointers. Is connected to Such a knowledge base system 1 is realized on a computer system such as a so-called workstation. The knowledge base 11, for example, is developed on a hard disk device, a floppy disk, or main memory,
Inference control and other processing operations of the inference engine 12 are performed by a central processing unit and the like, input of various commands and data by an operator is performed from a keyboard or a mouse, and the generated schedule draft is displayed on a screen of a display device. Displayed or printed by a printer device. Also, it is connected to another computer system, a host, a file device, or the like through an appropriate communication line, and various commands and data are exchanged. Next, based on a flowchart, the switching CHG management method in the scheduling process will be described in more detail. In the scheduling process, a plurality of candidate schedules are created by the schedule creation processing unit 23, the created schedules are evaluated, an optimal plan is determined from the schedules, and the determined schedule is checked. After the completion of the scheduling process, the operator corrects the proposed schedule as necessary. The scheduling process is executed for each order issued with information on work requirements such as the name of the product to be manufactured, its quantity, and delivery date. In other words, a schedule is created by determining the equipment, machine, date and time (real time), etc., for executing each order (for example, manufacturing a product). Normally, the scheduling process is collectively executed simultaneously for a plurality of orders within an appropriate period, and a schedule table is created based on a plurality of schedule plans obtained for each order. Now, the time series table 25a shown in FIG. 6 has already been created by the scheduling process, and in response to this, it is determined whether the work “B” can be arranged from the schedule creation processing unit 23 to the switching management unit 24. Suppose there is a confirmation instruction. In the confirmation instruction, the arrangement position of the operation “B” is designated in real time or by a relative positional relationship with respect to another already-arranged operation WRK. Here, it is assumed that there is an instruction to arrange after work “C”. Also,
In operation "B", the product is manufactured at 80 tons. FIG. 8 is a flowchart showing a processing operation by the work arrangement management unit 32. First, with respect to the position of the work WRK whose placement has been instructed, the names of the works arranged before and after the position and the times thereof are obtained (step # 11). Here, it is required that the tasks “C” and “A” are arranged before and after, respectively, and that the times are 1 to 4 o'clock and 12 to 16 o'clock, respectively. Next, the time t1 between the previous and next work WRK
Is calculated (step # 12). Here, assuming that 12-4 = 8, t1 = 8 hours is obtained.
In this case, the switching CH arranged after the operation “C”
G1 does not need to be considered since it becomes unnecessary when work “B” is arranged. For the work WRK whose arrangement is specified, the time t2 required for the work WRK is obtained (step # 13). Here, with reference to the work standard information 22, the production capacity MAB in the work “B” is 20 t / H, so that 80 ÷ 20 = 4, and t2 = 4 hours is obtained. The time t1 and the time t2 are compared, and it is determined whether the time t1 is smaller (step # 14). Here, since the time t1 is longer, the result is NO, and the process proceeds to step # 15. The switching times t3 and t4 for the switching CHG required before and after the WRK whose arrangement is designated are calculated (step # 15). The time t1 is compared with the time (t2 + t3 + t4), and it is determined whether the time t1 is smaller (step # 16). If the determination in step # 16 is no, the arrangement is possible, and a response to that effect is sent to the schedule creation processing unit 23 (step # 17). On the other hand, if the determinations in steps # 14 and # 16 are YES, the arrangement is impossible, and a response to that effect is sent to the schedule creation processing unit 23 (step # 18). FIG. 9 is a flowchart showing the processing operation by the switching time calculation unit 31. First, the work whose arrangement is specified by the identification flag FG
It is determined whether the WRK is a production type or a resource type (Step # 2)
1). If it is a production type, a production type switching time matrix for reference is selected (step # 22), and times t3 and t4 are directly obtained from the matrix (step # 2).
3). The matrix referred to here is, for example, the work switching time table 20 shown in FIG. The obtained times t3 and t4 are returned to the original routine (step # 24). If it is of the resource type, the resource RES to be used is determined with reference to the work standard information 22 (step # 25). Here, the operation “B” is a resource type, and the resource to be used is
As the RES, it is obtained that the model machine is “Y” and the cutter is “P”. For subsequent calculations, the times t3 and t4 are set to "0" (step # 26). Switching time t3a, t before and after for each resource RES
4a is obtained (step # 27). Here, first, regarding the molding machine, since “Z” and “Χ” are used in the operations “C” and “A”, respectively, the respective switching times from “Z” to “Y” and from “Y” to “Χ”
As for t3a, t3a = 1 hour and time t4a = 3 hours are obtained with reference to the resource switching time table 21a in FIG. Next, it is determined whether each of the obtained times t3a and t4a is longer than the times t3 and t4 (step # 28). If the times t3a and t4a are longer, the times t3a and t4a are substituted into the times t3 and t4, respectively. (Step # 29). Step # 27 and subsequent steps are repeated for all the resources RES (step # 30), and when this is completed (YES in step # 30), the finally obtained times t3 and t4 are returned (step # 24). Here, further regarding the cutter, referring to the resource switching time table 21b of FIG. 4, t3a = 1 hour, t4a =
Zero hours are obtained. As a result, t3 = 1 hour and t4 = 3 hours, which are the larger values, are returned. Then, in step # 16 of FIG. 8, the times t2, t3, t4
Is 4 + 1 + 3 = 8, which is equal to the time t1 = 8, so that the schedule creation processing unit 23 is replied to the schedule creation processing unit 23 at step # 17. In response to this, when there is a placement instruction instructing placement from the schedule creation processing unit 23, the placement unit 33 updates the time series table 25, and as shown in the time series table 25b of FIG.
And the switching CHG2,3 are arranged between the operation “C” and the operation “A”. According to the above-described embodiment, for the new arrangement state of the work WRK, such as the arrangement or deletion of the work WRK, the presence or absence of the switching CHG, and when the switching CHG is required, the switching time TCH is calculated, and the arrangement is performed accordingly. It is automatically determined whether or not is possible. As described above, since the switching CHGs are unifiedly managed, in the scheduling process, the operator need only be conscious of the work WRK to be scheduled without being conscious of the switching CHGs. Processing efficiency can be improved, and productivity in system construction can be improved. In the above embodiment, the case where the work WRK is arranged between the front and rear work WRK has been described, but the present invention can be applied to a case where the work WRK is arranged after a certain work WRK or only before a certain work WRK. . Also, the case where the resource RES is a mold and a cutter has been described, but the present invention can be applied to other machines, equipment, tools, personnel, and the like. Further, the processing contents of the switching time calculation unit 31 and the work arrangement management unit 32, the storage location or realization location of these, the identification flag FG or the time series table 25, or the knowledge base 11, the inference engine
12, the working memory 13, the configuration of the knowledge base system 1, the processing content, and the like can be variously changed in addition to the above.

【The invention's effect】

ADVANTAGE OF THE INVENTION According to this invention, the switching which arises between work can be managed uniformly, and the scheduling process can be made more efficient.

[Brief description of the drawings]

FIG. 1 is a block diagram of a knowledge base system according to the present invention, FIG. 2 is a diagram showing an example of a work switching time table, FIG. 3 is a diagram showing an example of a resource switching time table for a model machine, FIG. Is a diagram showing an example of a resource switching time table for a cutter, FIG. 5 is a diagram showing an example of work standard information, FIGS. 6 and 7 are diagrams showing an example of a time series table, and FIG. FIG. 9 is a flowchart showing a processing operation by the management unit, and FIG. 9 is a flowchart showing a processing operation by the switching time calculation unit. In the figure, 1 is a knowledge base system (scheduling device), 20 is a work switching time table, 21 is a resource switching time table, 22 is work standard information, 31 is a switching time calculation unit (switching time calculation means), and 32 is work allocation Management unit (work allocation management means), WRK is work, RES is resource, TCH is switching time, and FG is identification flag.

Claims (2)

(57) [Claims] 1. Work to be scheduled (WRK)
A work switching time table (20) that stores data on a switching time (TCH) from one work (WRK) to another work (WRK), The resource switching time table (21) that stores data on the switching time (TCH) from one resource (RES) to another resource (RES) and the resources (RES) required to perform the work (WRK) are shown. The work standard information (22) and are provided, and when instructing the placement of one work (WRK), the work (WK)
Set the identification flag (FG) for the type of (RK), and switch the time from another operation (WRK) that is arranged immediately before the operation (WRK) to the operation (WRK) in chronological order. TCH) and the work immediately following the work (WRK) (WR
The switching time (TCH) to K) is determined by referring to either the work switching time table (20) or the resource switching time table (21) according to the identification flag (FG). ) And the switching time (TC
H) is the sum of the immediately preceding operation (WRK) and the immediately following operation (WRK)
A task (WRK) that cannot be arranged when the time is longer than the time between the two. 2. Work to be scheduled (WRK)
A work switching time table (20) which stores data on a switching time (TCH) from one work (WRK) to another work (WRK), and Resource switching time table (21) that stores data on the switching time (TCH) from resource (RES) to another resource (RES), and work indicating the resources (RES) required to perform the work (WRK) When the standard information (22) and the placement of the work (WRK) are instructed, the work (WRK)
The work switching time table (20) or the resource switching time table (21) is referred to according to the identification flag (FG) for the type of K), and the work (WRK) is chronologically referenced. Other work (WRK) arranged in series
A switching time calculation means (31) for determining a switching time (TCH) from the operation (WRK) to the operation (WRK) and / or a switching time (TCH) from the operation (WRK) to the immediately succeeding operation (WRK); WRK), the switching time (TC
H) and a work arrangement management means (32) for comparing the time between the immediately preceding work (WRK) and the immediately succeeding work (WRK) to determine whether the work (WRK) can be arranged. A scheduling device comprising: