JP2021170239A - Apparatus, method and program for assisting process design - Google Patents

Abstract

To provide an apparatus for assisting process design capable of producing an operation sequence helpful in designing a production process.SOLUTION: An apparatus for assisting a process design in the present invention is adapted to assist a design of a production process to obtain an object through a plurality of operations. The apparatus for assisting a process design comprises: temporary candidate generating means that generates a temporary candidate that an operation sequence as an order to perform each operation is randomly supposed; matching means that makes the temporary candidate as a matching candidate that the temporary candidate is matched to an operation restriction that is a scheduled order between particular operations; and output means that outputs an evaluation value of the matching candidate based on a predetermined indicator. The temporary candidate consists of an operation sequence set by utilizing, for example, a pseudo-random number order. According to the present invention, a matching candidate matched to an operation restriction can be considered on the basis of a number of temporary candidates randomly set by a system. For this reason, there is a room that a process design more excellent in productivity is to be made freely from the conventional framework.SELECTED DRAWING: Figure 7

Description

The present invention relates to a support device or the like useful for designing a production process.

Product productivity is highly dependent on the production process. The production process usually consists of a plurality of processes, and each process consists of a plurality of operations. When the number of operations and the number of processes becomes enormous, it becomes difficult to design an efficient production process only by humans. Therefore, an apparatus (simulation apparatus) that supports the design of a production process has been proposed, and for example, there is a description related to the following patent documents.

Japanese Unexamined Patent Publication No. 2003-15723

In Patent Document 1, a process design support device (process design support device) in which various operations are performed for each stage (process) arranged in series along a line through which a work is conveyed, such as an automobile production line (assembly line). We are proposing a work allocation planning support device that uses a visual display screen). Specifically, the support device displays the total time of working time and walking time allocated to each worker (that is, each process) on the screen so that the process designer can change them on the screen. I have to. With such a support device, the load on the process designer can be reduced.

However, in the process design using the conventional support device, the existing process, work order, etc. are only partially optimized within the competence (experience, knowledge, etc.) of the person in charge.

An object of the present invention is to provide a support device or the like that enables an overall suitable process design by examining a new process that is not bound by an existing process.

As a result of diligent research to solve this problem, the present inventor has made each work based on a plurality of work sequences (provisional candidates) arbitrarily generated without being bound by the existing process or the ability of the designer when designing the process. The idea was to examine the work order (matching candidate) generated so as to satisfy the work restrictions related to the order between them, and the generation of such a work order was realized. By developing such results, the present invention described below has been completed.

<< Process design support device >>
(1) The present invention is a process design support device that supports the design of a production process for obtaining a target product through a plurality of operations, and generates a tentative candidate that randomly assumes a work order that is the order in which each operation is performed. Outputs the evaluation value of the matching candidate according to a predetermined index, the provisional candidate generation means for making the provisional candidate, the matching means for making the provisional candidate a matching candidate matched to the work constraint in the order scheduled between specific works, and the matching candidate. It is a process design support device including an output means for performing.

(2) According to the process design support device (simply referred to as "support device") of the present invention, first, a random (mechanical, arbitrary, provisional) work sequence (provisional candidate) is performed by the device (system, computer). ) Is temporarily set. Next, based on the provisional candidate, a work order (matching candidate) that matches the work constraint is obtained. Then, the matching candidate is evaluated by a predetermined index. The evaluated matching candidates (work order) can be utilized for subsequent process design (division of work according to the work order into each process, etc.).

In this way, according to the support device of the present invention, it is possible to study a work sequence and a process based on the work sequence that are not bound by the conventional framework (existing process, ability of the designer, etc.). As a result, it is possible not only to shorten the time required for process organization (lead time) and reduce the burden on the process designer, but also to create suitable and optimal work sequences and process groups that contribute to the improvement of productivity. obtain.

<< Process design support method >>
The present invention can also be grasped as a process design support method (simply referred to as a "support method"). That is, the present invention is a process design support method that supports the design of a production process for obtaining a target product through a plurality of operations, and generates a tentative candidate that randomly assumes the work order in which each operation is performed. Output that outputs the candidate generation step, the matching step that makes the provisional candidate a matching candidate that is matched to the work constraint in the order scheduled between specific works, and the evaluation value of the matching candidate by a predetermined index. It may be a process design support method including steps.

<< Process design support program >>
The present invention may also be understood as a process design support program (simply referred to as a "support program") that causes a computer to execute each of the above steps.

"others"
(1) The process design support according to the present invention is realized by a computer and a program executed by the computer. A computer usually includes a calculation unit (CPU or the like), a storage unit (memory or the like), a display unit (display or the like), and an input unit (keyboard or the like). The computer may be isolated or linked by a communication network (for example, a server and a client).

(2) Each means or each step according to the present invention is usually configured or realized by a program executed by the computer. The "-means" and "-step" referred to in the present specification can be paraphrased with each other according to the object of the present invention (invention of a product or invention of a method). "-Means" may be paraphrased as "-department", and "support device" may be paraphrased as "support system". In addition, a support device composed of a plurality of linked devices (computer, storage device (server, etc.), terminal, etc.) is also grasped as a "support system".

(3) Process design support is provided by numerical analysis (simulation) of a virtual work (analysis model) set (converted into data) corresponding to an actual work (simply referred to as "actual work"). Therefore, unless otherwise specified, the "work" referred to in the present specification consists of a group of data reflecting the actual work.

"Work" (data) includes, for example, an identifier (number, code, etc.) assigned to each actual work, and features of the actual work (work content, work position, work time, jigs / tools to be used, etc./" It is called "work information") and has work restrictions. The work constraint is information necessary for generating a work order, for example, information on the predecessor-sequential relationship of the work order (order constraint), information on the continuous relationship of the order (continuous constraint), and discontinuity of the order (continuous constraint). There is information about (continuity not possible) (intermittent constraint), information that brings the order position closer to each other (neighborhood constraint), and the like. Work constraints may be set for all tasks or for some of the tasks, at least between specific tasks or for a group of tasks. The "work order" referred to in the present specification includes an order of individual work (single work) and an order of grouped work groups (including a case consisting of one work) that satisfy continuous constraints.

(4) The target object for process design may be an intermediate product (semi-finished product or work-in-process) as well as a final product. Regardless of the content of each work, in addition to assembling and joining parts, manufacturing of raw materials may be performed.

It is a table exemplifying work information, order constraint and continuous constraint. It is a table which exemplifies conversion from work ID to group ID. It is a table exemplifying the provisional candidate of the work order generated based on the group ID. It is a matrix showing an example of a tentative candidate. It is a matrix that aggregates similar work based on the provisional candidates. It is a dendrogram showing the aggregation process of the aggregated similar work. It is a matrix showing matching candidates that aggregates the similar works. It is a matrix showing matching candidates that did not aggregate the similar work. This is an example of a list showing the work order and evaluation values of matching candidates. It is a matrix in which the work order and the evaluation value are displayed for each matching candidate. It is a flowchart which exemplifies the generation procedure of the work order.

One or more components arbitrarily selected from the present specification may be added to the above-described components of the present invention. The contents described in this specification appropriately apply not only to the process design support device (system) but also to the support method, support program, and the like.

《Temporary candidate generation》
Temporary candidate generation (the notation of "means" or "step" is omitted; the same applies hereinafter) generates a provisional candidate in which the work order in which each work is performed is randomly set. The tentative candidate may be a work order generated according to a certain rule (algorithm, etc.). For example, tentative candidates may be generated based on optimization algorithms such as Genetic Algorithms (GA). The randomly assumed work sequence may be part of the entire work sequence.

The tentative candidate may consist of, for example, a work order set based on a pseudo-random number sequence. When there are m works, the work order is m! (Factial) There can be streets. As the number of works (m) increases, the number (m!) Becomes enormous, and it becomes impossible to consider all the work sequences.

Therefore, for example, if each work to which an identifier (ID / number) is assigned is arranged based on a pseudo-random number sequence (simply referred to as a "random number sequence"), a desired number of random work order (provisional candidates) groups are generated. It becomes possible to do. The array of work order (ID) may be generated by a random number sequence excluding duplicates. The work order of the tentative candidates does not necessarily have to be the random number sequence itself.

《Alignment》
In the harmonization, the work order of the provisional candidate is set to the work order (alignment candidate) that is consistent with the order (work constraint) scheduled between specific works. Typical examples of work constraints are the order constraint, which is the predecessor relationship of the order scheduled between specific tasks, the continuous constraint, which is the continuous relationship of the order scheduled for a specific work group, and the work constraints. There is a discontinuous relationship that prohibits continuity (intermittent constraint), and the order of specific work intervals / work groups is kept within a predetermined range (neighborhood constraint).

Consistency candidates are particularly good at generating work sequences in which similar works are close together. The similarity of work is determined based on, for example, the information distance obtained from the information (work information or work constraint) given / set to each work. In this case, the work in which the information distance is within the predetermined range is a similar work. Work in which the information distance is within a predetermined range (simply referred to as “close work”) is, for example, work in which jigs, tools, target parts, and the like to be used are common or related. The work order in which related similar works are aggregated in this way enables process design that can improve productivity.

It is preferable that the work groups related to the continuous constraint are collectively aligned and a matching candidate (work order) is generated. As a result, efficient matching can be achieved even when there is a continuous group of work. The batch-processed work group may be decomposed (ungrouped) into individual works after the work order is determined. As a result, a work order (alignment candidate) in which each work satisfies the continuous constraint can be obtained.

"output"
It is preferable that the evaluation value based on a predetermined index is output for the matching candidate. The evaluation value may be shown together with the identifier of the matching candidate or its work order. Further, sorted matching candidates may be displayed based on the evaluation value.

The evaluation value can be calculated based on, for example, each work information (work place, work time, etc.). As an example of the evaluation value, there is a total distance that the worker moves along the work order of the matching candidate (for example, the sum of the movement distance associated with the performance of one work and the movement distance between different work places).

In order to explain the process design support device (system) according to the present invention, the present invention will be described more concretely while showing a virtual case (model).

"Device configuration"
The support device (system) according to this embodiment is composed of a computer and a program executed on the computer. A computer is an arithmetic unit (CPU, etc.), a storage unit (storage device such as ROM, RAM, HDD, SSD, etc.) that temporarily or long-term stores data and programs, a display unit (display, etc.), and an input / output unit (keyboard). , Mouse, communication interface, etc.). The computer may be a dedicated machine or a general-purpose machine (personal computer (PC) or the like). The program is temporarily or long-term installed on the computer to perform each step of the process design support method described below. Each means according to the present invention is composed of a program portion that executes each step.

"model"
As an example of process design, it is assumed that a production line (production process) consisting of a huge number of work elements (simply referred to as "work") is designed, such as an automobile assembly line. The production line may be only a series (straight line) line, but usually, a large number of lines (tributary lines) branched on the upstream side are merged on the downstream side and integrated into the main line (main stream line). I often go. Each line is divided into one or more processes, and one or more operations (referred to as "work group") are assigned to each process.

When considering the work (group) to be assigned to each process of a certain line, if all the work to be processed on that line is given as a work order (candidate) arranged in order in chronological order, the process design is performed. It can be done efficiently. This is because each process is determined only by dividing a series of work sequences based on a predetermined index (for example, required time of one process, tact time, workability, etc.).

Based on such a situation, in this embodiment, a method (device, system) for supporting the generation of the work sequence required for process design will be described. In addition, in order to facilitate understanding and for convenience of explanation, the number of tasks to be examined was intentionally reduced in this embodiment.

《Generation of work order》
FIG. 7 shows a processing procedure (flow chart) for generating a work sequence candidate that is meaningful for process design. Each procedure (step) will be described in detail according to the flowchart.

(1) Information acquisition (input)
In step S1, various information necessary for generating the work order is acquired from the database. Specifically, the work information is input in step S11. The work information is, for example, as shown in Table 1A of FIG. 1, a work content (work name), a time required for a single work (work time), a work position (XY coordinates), and the like.

A number (ID), which is an identifier, was assigned to each work. When distinguishing each work, it is written (named) as "work 1", "work 2", etc. based on the ID. This point also applies to the groups, candidates, etc. described below.

The work constraint is input in step S12. Work constraints are, for example, order constraints and continuous constraints. As shown in Table 1B of FIG. 1, the order constraint is a first-to-back relationship of the order in which each work is performed. For example, work 1 is associated with preceding work 7. In this case, work 1 is in the order after work 7. In this embodiment, the case where the order constraint is provided only for a part of the work, not all the work, is illustrated.

As shown in Table 1C of FIG. 1, the continuous constraint is a continuous relationship in the order in which each operation is performed. An identifier is assigned to each continuous constraint. For example, the first continuous constraint (simply referred to as "continuous 1") stipulates that work 6, work 8, work 9 and work 10 are continuously performed in that order.

In step S13, the number of work sequence candidates (n: natural number) desired to be presented by the system of this embodiment is input. Steps S11 to S13 are collectively referred to as step S1.

As shown in step S8 described later, the processes of steps S3 to S8 are repeated until the number of presented work orders (i: natural number / i is incremented) becomes the number of candidates (n).

(2) Grouping of continuous work In step S2, the work related to the continuous constraint is grouped so that the work can be collectively processed. For example, each work (6, 8, 9, 10) related to one continuous constraint (continuous 1) described above is designated as one group 1. In this embodiment, a group ID is also assigned to a work ID having no continuous constraint. An example thereof is shown in Table 2 of FIG. As a result, work 6, 8, 9, and 10 are group 6, work 1 is group 1, work 2 is group 2, and all the work is uniformly processed by the group ID.

(3) Generation of provisional candidates for work order In step S3, provisional candidates consisting of work orders randomly arranged (at random) for each work are generated based on the group ID. An identifier (candidate ID) was also given to each candidate obtained in this way. A pseudo-random sequence was used for the random array of group IDs.

An example of the provisional candidates thus obtained is shown in Table 3 of FIG. As is clear from Table 3, the work order of the provisional candidates basically does not satisfy the order constraint (Table 1B).

(4) Aggregation of similar works In step S4, each work presented as a provisional candidate is aggregated based on the information distance belonging to them. As an example, a case where similar work is aggregated based on an order constraint will be described. Create a matrix in which the group IDs (work order) shown in Candidate 1 in Table 3 are shown in rows and columns, respectively. The order constraint (predecessor group ID converted from the predecessor work ID) shown in Table 1B is added to the matrix as the flag "1". This gives the matrix shown in Table 4A of FIG. 4A.

Looking at the group ID attached to the row, it can be seen that, for example, group 7 (work 7) exists as a preceding work in group 1 (work 1). However, as can be seen from Table 4A, "1" indicating the order constraint is in a state of being dispersed over the entire matrix.

When the matrix shown in Table 4A is subjected to a hierarchical clustering process, as shown in Table 4B of FIG. 4B, "1" indicating an order constraint is arranged near the diagonal line of the matrix. Then, each work (group) is aggregated on the upper left side and the lower right side of the matrix. As a result, all the work is classified (clustered) into a set consisting of groups 1, 2, and 7 and a set consisting of groups 5, 6, 3, and 4. As is clear from Table 4B, the order (row order) of the group IDs does not satisfy the order constraint (Table 1B) even at this stage.

The matrix generation process (hierarchical clustering process) shown in Table 4B is shown in FIG. 4C as a dendrogram. The lower part of FIG. 4C shows that the information distance is short (the degree of similarity is large).

Here, for the sake of simplicity, similar work is summarized based on the order constraint, but it may be done based on information on the work position, the type and material of the part to be worked, the jig and tool to be used, etc. good. Of course, the aggregation of similar work may be performed by combining a plurality of types of information. Further, a method other than hierarchical clustering may be used for aggregating similar works. Further, the process designer may select the aggregated input information and method by the preliminary experiment.

In any case, by aggregating similar works in advance, similar works (for example, works using the same tool) are put together, and it becomes easy to propose a work order (candidate) having excellent productivity.

(5) Consistency of work constraints (satisfaction of order constraints)
In step S5, the similar works are rearranged so as to satisfy the work constraint (order constraint) based on the aggregated work sequence candidates. For example, when the work order shown in Table 4B is rearranged so as to satisfy the order constraint shown in Table 1B, the work order (matching candidate) shown in Table 5A can be obtained. Specifically, only the order of the group IDs that violated the order constraint is sorted in order from the first row of the matrix. Conversely, the order of group IDs that do not violate the order constraint is maintained. By such a partitioning process, a matrix (Table 5A) in which "1" indicating an order constraint is displayed only below the diagonal line extending from left to right is obtained.

The matching candidates may be generated by focusing only on the satisfaction of the order constraint, or of other evaluation indexes (for example, the neighborhood constraint: "Perform work A and work B at positions close to each other"). It may be done in consideration of satisfaction and the like. Incidentally, if only the alignment of the order constraint is performed without aggregating the similar operations described above, for example, the operation order (alignment candidate) as shown in Table 5B of FIG. 5B can be obtained. It is also possible to consider such a work sequence. However, in general, it is difficult to obtain a highly efficient production process from a work order in which the orders of similar works are separated.

(6) Canceling the group of continuous work In step S6, the group of continuous work handled collectively is released. For example, the group 6 is released and returned to work 6, 8, 9, and 10. In this way, the work order that is a matching candidate is obtained for each work as a whole (see Table 6 in FIG. 6A).

(7) Output of matching candidate and evaluation value In step S7, the entire work order as the matching candidate and the evaluation value of the matching candidate are output. An example is shown in Table 6.

The evaluation value is, for example, the total movement distance (total movement distance) between adjacent works constituting the matching candidate. The moving distance between adjacent works can be obtained from, for example, the work position (coordinates) included in the work information (Table 1A).

The output formation may be in a list format as shown in Table 6 or in an individual format for each matching candidate as shown in FIG. 6B. In the matrix shown in FIG. 6B, each working position is visualized by color-coded display. In addition, the order distance from the preceding work is also visualized and displayed. In this case, the closer it is to the diagonal line, the closer the order distance to the preceding work is. The absence of a 1 in the upper right corner of the diagonal indicates that there is no violation of the order constraint.

(8) Number of Candidates Each of the above steps is repeated until the work order of the set number of candidates (n) is obtained (step S8). As a result, for example, the list shown in Table 6 is completed.

The plurality of matching candidates may be sorted and displayed based on the evaluation value, for example. Further, when there are a large number of matching candidates, representative matching candidates may be shown for each group by grouping them from a predetermined viewpoint.

By utilizing the large number of work sequences presented as described above, it becomes possible to efficiently design and examine the optimum or near-optimal production process as a whole.

Claims (6)

It is a process design support device that supports the design of a production process that obtains a target product through multiple operations.
A tentative candidate generation means that generates a tentative candidate that randomly assumes a work order that is the order in which each work is performed, and
A matching means that makes the provisional candidate a matching candidate that is matched to work constraints that are in the order scheduled between specific works, and
An output means for outputting the evaluation value of the matching candidate according to a predetermined index, and
Process design support device equipped with. The process design support device according to claim 1, wherein the tentative candidate has a work order set based on a pseudo-random number sequence. The work constraint includes an order constraint that is a pre-existing relationship of the order scheduled between specific tasks.
The process design support device according to claim 1 or 2, wherein the matching candidate has a work order in which similar work orders are brought close to each other. The work constraint includes a continuous constraint which is a continuous relationship of the order scheduled for a specific work group.
The process design support device according to any one of claims 1 to 3, wherein the matching means collectively handles a work group related to the continuous constraint. It is a process design support method that supports the design of a production process that obtains a target product through multiple operations.
A tentative candidate generation step that generates a tentative candidate that randomly assumes the work order that is the order in which each work is performed,
A matching step that makes the provisional candidate a matching candidate that is matched to work constraints that are in the order scheduled between specific tasks, and a matching step.
An output step that outputs the evaluation value of the matching candidate according to a predetermined index, and
Process design support method equipped with. A process design support program that causes a computer to execute each step according to claim 5.

Images