JP2865520B2 - Lot product allocation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a single lot product for a certain width of material, such as a case where lots of orders are combined to make a lot product such as a plan for cutting paper, cloth, and cellophane. The present invention relates to a lot product allocating apparatus for generating a combination pattern for allocating a width of a product.

[0002]

2. Description of the Related Art Conventionally, for example, when making a lot product of cellophane, for example, a wide product (for example, 510
0mm) roll, a plurality of cutters are arranged at a position corresponding to the order width of the lot product (400mm to 1220mm) with respect to the roll of the semi-finished product, and lot products of various widths are pulled out from the roll while the semi-finished product is pulled out. Are cut at the same time.

[0003] By the way, according to such a cutting method, a semi-finished edge is formed at the end of the semi-finished product. Therefore, according to the width and the number of the ordered lot product, the width of the order is smaller than the width of the semi-finished product (material width). There is a need to reduce the waste of materials by allocating well, and it is required to find a combination pattern that allocates the width of this order.

In order to obtain such a combination pattern, there are a method of solving by an expert, a linear programming method, and a method of an expert system using rules. A method of checking all combinations is also conceivable.

[0005]

However, in the method of solving by an expert, there is a problem that it takes time since a plurality of combinations are conceivable, and that a plan between the experts tends to be inadequate.

In the linear programming, it is necessary to solve a simultaneous equation having a degree of freedom corresponding to the number of order types. If the number of order types increases, the calculation takes a long time. Furthermore, since all the patterns cannot be determined at once by the linear programming, it is necessary to repeat a time-consuming calculation until all the orders are satisfied, and furthermore, a calculation time is required.

[0007] The method using rules is to find some rules from an expert's method and to imitate them on a machine. However, exceptions are likely to appear in the rules. Rules are required. Note that the method of checking all combinations may not be able to be solved in a feasible time because the number of combinations increases explosively.

SUMMARY OF THE INVENTION It is an object of the present invention to generate a combination pattern for assigning the width of a single lot product to the material width of a fixed width material at a high speed to facilitate a cutting plan and the like.

[0009]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, a lot product allocating apparatus according to the present invention cuts out single articles having different widths from a material having a fixed width and prepares a plurality of pieces according to the width of the cut single articles. A lot product allocating apparatus for generating a combination pattern for allocating the width of a single product of the lot product to the material width of the material in order to manufacture various types of lot products , as shown in FIG. A lot information input means A for inputting lot information including the width of each lot product and the number of single products of each lot product; a lot information storage means B for storing the input lot information; For the selected group consisting of single items, a combination pattern consisting of a set of single items such that the sum of the widths of the selected single items is less than or equal to the material width while randomly selecting the single items, , A combination pattern storage means D for storing the combination patterns generated by the combination pattern generation means C, a sum of the widths of the single products in the combination pattern, and the material width. And a combination pattern selecting means E for selecting a recombination pattern from the combination patterns in the combination pattern storage means D such that a combination pattern having a larger difference is selected with a higher probability.
And output means F for outputting information on the combination pattern stored in the combination pattern storage means D, and combination pattern evaluation means G for detecting that the difference has entered an allowable range and evaluating the combination pattern. The combination pattern generation means C generates a combination pattern for the single item of the lot information stored in the lot information storage means B, stores the combination pattern in the combination pattern storage means D, and then stores the combination pattern in the combination pattern storage means D. The operation of selecting a re-combination pattern by the combination pattern selecting means E, and the generation of a combination pattern by the combination pattern generating means C for all the single products of the re-combination pattern to the combination pattern not selected by the combination pattern selecting means E The operation of joining and storing in the combination pattern storage means D is repeated. By repeating the above operation while evaluating the combination pattern by the combination pattern evaluation means G, information on the combination pattern evaluated by the combination pattern evaluation means G is output to the output means F. .

[0010]

FIG. 2 is a diagram for explaining the principle of the present invention. In FIG. 2, "、", "●" and "◎" indicate one set of combination patterns, and "●" indicates combination pattern selection means E. The selected recombination pattern, “「 ”indicates the combination pattern generated by the combination pattern generation means C for all the single products of the recombination pattern. Also, ~ indicates segments corresponding to the difference between the sum of the width of the single item in each combination pattern width of material, order selection probability as will color is high.

First, when a combination pattern is generated for a single item of the lot information stored in the lot information storage means B and stored in the combination pattern storage means D, each of the combination patterns is divided into respective sections as shown in FIG. Distributed approximately equally to When a re-combination pattern is selected by the combination pattern selection means E from the combination pattern of the combination pattern storage means D, for example, a large number of combination patterns are sequentially selected as re-combination patterns in a section such as "●" in the drawing. Become.

Next, when the recombination pattern is separated as shown in (II) and a combination pattern is generated by the combination pattern generating means C for all the single products of the recombination pattern, each combination pattern is represented by (III) in FIG. As shown in “◎”, the distribution is substantially equally distributed to each section. Therefore, when the regenerated combination pattern "◎" is added to the combination pattern not selected by the combination pattern selection means E as shown in (I ') and stored in the combination pattern storage means D, (IV) in FIG. become that way.

When this operation is repeated and a re-combination pattern is selected by the combination pattern selection means E from the combination pattern in the combination pattern storage means D, for example, the re-combination pattern is sequentially divided into, for example, "●" in FIG. Many combination patterns are selected as re-combination patterns.

Therefore, when the above-described repetitive operation is performed, the number of combination patterns having a small difference between the material width and the sum of the widths of the individual products in each combination pattern increases, and the number of combination patterns having a large difference decreases. Therefore, when the combination pattern evaluating means G outputs the information of the combination pattern to the output means F when the largest difference becomes equal to or less than a predetermined threshold value, for example, a combination pattern having a small difference, that is, a combination pattern having a small amount of material is automatically generated. Is obtained.

[0015]

DETAILED DESCRIPTION FIG. 3 is a block diagram of a lot products allocation apparatus of the present onset AkiraMinoru施例. The apparatus of this embodiment is that consists of a control program supplied from the personal computer and the floppy disk.

The CPU 1 reads data of a control program from a floppy disk set in a floppy disk driver 3 based on a system program stored in a ROM 2 and stores the control program in a RAM.
4, and controls the generation of the combination pattern using the working area of the RAM 4 based on the control program stored in the RAM 4. Then, the generated combination pattern information is displayed on the display 5 in a predetermined manner.
And print out to the printer 6.

Further, in the case of cellophane, data on the material width of a material such as a semi-finished product, threshold data for evaluating the difference between the material width and the total length of the combination pattern (hereinafter, referred to as “remainder” in this embodiment). The selection probability data for selecting each combination pattern as a re-combination pattern with a probability corresponding to the remainder is read out from the floppy disk together with the control program and stored in the RAM 4.

The material width data, surplus threshold data, surplus division data, and selection probability data are prepared as a plurality of preset values according to the type of order, such as paper, cloth, and cellophane. Have been.

On the other hand, the order data of the width and the number of the lot products according to the order are input by operating the keyboard 7, and the CPU 1
Stores the order data input from the keyboard 7 in the RAM 4. Further, by operating the keyboard 7 or the like, data of the material width, data of the remaining threshold value, data of the remaining division, and data of the selection probability are selected. These data may be changed by operating the keyboard 7.

Assume that a lot product having a width and a number as shown in FIG. 5 is ordered.
The case of allocating to mm material (semi-finished product) will be described. In this embodiment, the sections for dividing each combination pattern according to the remainder are divided into seven sections as shown in Table 1 below.
, And a selection probability indicating a probability that a combination pattern is selected from the section for replanning is set for each section.

[0021]

[Table 1]

When generating a combination pattern, first,
As shown in FIG. 6, the width of the selected single item is changed as shown in FIG. 6 by randomly selecting any single item from the selection group (single item set as a whole) shown in FIG. 5 and excluding the single item from the selection group. Are sequentially added. Then, as shown by the broken line in FIG. 6, when the added value exceeds the material width (5100 mm),
The last single item is returned to the selected group, and the previous single item group is set as one combination pattern.

The same processing is repeated until there is no single product from the selected group, and a plurality of combination patterns are generated. Then, the remainder for each combination pattern is obtained, and each combination pattern is allocated to the following categories according to the remainder.

Next, the number of combination patterns corresponding to the selection probability is selected from each section as a re-combination pattern, separated from the remaining combination patterns, and all the single items included in the re-combination pattern are selected as a selection group, as described above. To find the combination pattern. Then, the combination pattern obtained based on the recombination pattern is allocated to the following categories according to the remainder, and is added to the remaining combination patterns not selected as the recombination pattern (subscription).

In this embodiment, when at least the second largest remainder of the generated combination patterns becomes smaller than the target value (for example, 200 mm), the calculation is terminated, and all the combination patterns at this time are terminated. Is displayed,
Output.

FIG. 4 is a flowchart of the control program. When the control program is started, first, in step S1, initialization is performed, and in step S2, order data input processing is performed. The order data input processing is performed by a known technique such as data input based on the operation of the keyboard 7, and the input data of the width and the number of the single item is stored in the RAM 4.

When the order data input processing is completed, in step S3, a combination pattern is generated for the input order data. In step S4, the total width of each combination pattern and the material width of the semi-finished product (this (5100 mm in the example), that is, a combination pattern evaluation process for obtaining a remainder.

Next, in step S5, it is determined whether or not at least the second largest remainder among the remainders of the combination pattern has become equal to or less than the target value. If the target value has been reached, the data of the combination pattern is deleted in step S10. The information is displayed on the display 5 and printed out to the printer 6, and the process ends.

If the target value has not been reached in step S5, a recombination pattern is selected in step S6 based on the selection probability according to the remainder, the number of the recombination patterns is stored, and the remaining patterns are stored. As the original pattern, in step S7, a combination pattern is generated for all single products of the recombination pattern.

Next, in step S8, it is determined whether or not the number of patterns of the currently generated combination pattern is greater than the number of patterns of the combination pattern selected in step S6, that is, whether or not the used material does not exceed the combination pattern. It is determined, and if it exceeds, the process returns to step S7 and the same process is performed. If it does not exceed, the combination pattern currently generated in step S9 is added to the original pattern, and the processes after step S4 are performed.

By the above processing, as described with reference to FIG. 2, the combination patterns with few remaining numbers are sequentially increased, and when they reach the target value, they are efficiently allocated, for example, as shown in FIG. A combined pattern (seven patterns in this example) is obtained.

In order to randomly generate a combination pattern, for example, the order data is assigned to each individual item using array variables and the like, variables are assigned, and the width of each individual item is stored. , Select the order by random number, etc., select the variables so that they do not overlap at random, and sequentially accumulate the width of the selected variables, and when the accumulated value exceeds the material width, the combination of the number of the previous variable As one combination pattern. In this case, in order to prevent the variable from being selected redundantly, a flag or the like may be set corresponding to each variable (single item).

In addition, storage of the generated combination pattern,
Data processing such as evaluation of a target value, separation of a re-combination pattern, and addition to an original combination pattern can be performed by conventionally known data processing.

The processing of the above embodiment does not need to perform a real number operation and may be an integer operation, so that the processing can be performed at high speed. Further, the configuration is simpler than that of the expert system, and the required storage area may be small. Furthermore, the solution can be obtained at a higher speed compared to expert judgment and linear programming.

Since a random number is used, when the result is not preferable, another solution can be obtained by starting over.
Note that the same result is obtained every time using a method that does not use random numbers. Therefore, when a combination that is not suitable for the nature of a program is used, it is necessary for a human to re-plan from the beginning.

By the way, in many make-to-stock productions, the number to be created is increased or decreased according to the number of materials that can be planned and used. However, since the apparatus of the present invention can be processed at high speed, Experts can interactively simulate changing these inputs (order data) by looking at the results and can concentrate on advanced planning as a whole.

[0037]

As described above, according to the present invention, based on the lot information including the width of each lot product and the number of single products of each lot product, selection of single products having a known width and number is performed.
For the selected group, select all of the single products in the selected group by selecting a single product randomly and selecting a combination pattern consisting of a set of single products such that the sum of the widths of the selected single products is equal to or less than the material width.
The sum of the widths of single items in the combination pattern that occurs only in
The re-combination pattern is selected with a probability according to the difference between the re-combination pattern and the material width so that a combination pattern having a larger difference has a higher selection probability, and a re-combination pattern is generated again for all single products of the re-combination pattern. Joining the combination pattern that was not selected as the combination pattern, evaluating the combination pattern by the difference while repeatedly performing the selection of the re-combination pattern based on such a probability and the generation of the combination pattern, and the evaluated combination pattern. Output information, such as when cutting lots of orders such as paper, cloth, cellophane, etc. to create lot products by combining many types of orders. A combination pattern for allocating a width can be generated at a high speed to easily perform a cutting plan and the like.

[Brief description of the drawings]

FIG. 1 is a basic configuration diagram of a lot product allocation apparatus of the present invention.

FIG. 2 is a diagram illustrating the principle of the present invention.

FIG. 3 is a block diagram of a lot product allocation apparatus of the present onset AkiraMinoru施例.

FIG. 4 is a flowchart of a control program according to the embodiment of the present invention.

FIG. 5 is a diagram illustrating an example of order data according to the embodiment of the present invention.

FIG. 6 is a diagram illustrating a process of generating a combination pattern according to the embodiment of the present invention.

FIG. 7 is a diagram illustrating an example of a processing result according to the embodiment of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... CPU, 2 ... ROM, 3 ... Floppy disk driver, 4 ... RAM, 5 ... Display, 6 ... Printer, 7 ... Keyboard.

Claims (1)

(57) [Claims] 1. A single product of said lot product with respect to the material width of said material, in order to cut out a single product having a different width from a material of a fixed width and to manufacture a plurality of kinds of lot products according to the width of said cut single product. A lot product allocating apparatus for generating a combination pattern for allocating a width of the lot product, wherein lot information input means for inputting lot information including the width of each lot product and the number of single products of each lot product; A lot information storage means for storing the lot information obtained, and a selection group consisting of single items of which width and number are known, wherein the single items are selected at random and the sum of the widths of the selected single items is equal to or less than the material width. Combination pattern generating means for generating a combination pattern consisting of a set by a number that selects all the single products of the selected group, and the combination pattern generating means A combination pattern storing means for storing the combination pattern, and a probability corresponding to the difference between the sum of the widths of the single items in the combination pattern and the material width, and a combination pattern having a larger difference is selected with a higher probability. Combination pattern selection means for selecting a re-combination pattern from the combination patterns in the combination pattern storage means; output means for outputting information on the combination patterns stored in the combination pattern storage means; And a combination pattern evaluation unit for evaluating the combination pattern by detecting that the combination pattern is generated by the combination pattern generation unit for the single item of the lot information stored in the lot information storage unit. Means, and then stored in the combination pattern storage means. An operation of selecting the re-combination pattern by the combination pattern selecting means for the combined pattern, and generating a combination pattern by the combination pattern generating means for all single products of the re-combination pattern and not selecting the combination pattern by the combination pattern selecting means The repetition of the operation of joining the combination pattern and storing the combination pattern in the combination pattern storage means is performed, and the above-described repetition operation is performed while evaluating the combination pattern by the combination pattern evaluation means. A lot product allocating device for outputting information of the selected combination pattern to the output means.