JP2010229510A - Treatment-order creation apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically determine a candidate of a treatment order in a treatment line for a plurality of strip-shaped members even when the plurality of the strip-shaped members to be treated include strip-shaped members that have not actually been treated in past years, and select an adequate candidate of the treatment order from the qualitative point of view. <P>SOLUTION: In the treatment-order creation apparatus, a treatment-order evaluation means 36 determines error levels ELt of treatment orders for N candidates PR of the treatment orders respectively, and a best-solution selection means 40 selects the candidate of which the error level of the treatment order is minimal out of a plurality of the candidates of the treatment order, as the best solution. The error level of the treatment order is calculated based on the attribute category to be considered for each treatment facility pre-stored in a connection evaluation table, a tolerance of the attribute difference of the connecting part, which is quantified step by step of each attribute category, and a weighting factor preset for each treatment facility. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention is applied to a processing line in which a plurality of strip members are connected and continuously processed by one or more processing facilities, and the best processing order candidate for the plurality of strip members processed in this processing line is created. The present invention relates to a processing order creation device.

For example, in a production line including processing equipment such as continuous annealing equipment and cold rolling equipment in an ironworks, it is required to continuously process a coil wound with steel plates having different dimensions and materials. Therefore, a plurality of coils to be continuously processed are sequentially connected to operate continuously.
By the way, when the dimension and material of a steel plate differ greatly between coils to be connected, it may be necessary to change manufacturing conditions such as a plate passing speed and a heating temperature in a processing facility in a production line. On the other hand, when the manufacturing conditions are greatly changed, various troubles may occur such as a quality defect occurring at the joint between the steel plate of the preceding coil (preceding steel plate) and the steel plate of the succeeding coil (subsequent steel plate). For this reason, acceptable reference values are determined for differences in dimensions, materials, and the like between the steel plates to be connected, and production plans are made so as not to connect coils that do not satisfy the criteria.

However, in order to make a production plan composed of many coils (steel plates) of several tens or more, the processing order must be determined so that all the steel plates satisfy the above-mentioned standard value. Such planning is very complicated.
As an apparatus (process order creation apparatus) for efficiently performing the process order planning work for a large number of steel plates as described above, the apparatus described in Patent Document 1 is known. The processing sequence creation device described in Patent Document 1 displays, for each attribute, the ease of connection between each band-shaped member based on the past processing results for each band-shaped member for a plurality of band-shaped members to be processed. Based on the connectivity table creation means for creating the connectivity table and the connectivity table for each attribute created by the connectivity table creation means, the range of strip members that can be processed first for each strip member is determined. The calculation means for obtaining the degree of freedom of connection in the front and the degree of freedom of connection in the rear representing the range of the band-shaped member that can be processed later, and the degree of freedom of connection in the front based on the degree of freedom of connection obtained by the means of calculation There is provided a selection means for preferentially selecting a belt-like member having a small size and a large degree of freedom of connection at the rear. Thereby, the processing order with respect to many strip | belt-shaped members is created automatically by preferentially selecting the strip | belt-shaped member with a small front connection freedom and a large back connection freedom preferentially.

JP-A-8-112607

  However, in the processing sequence creation device described in Patent Document 1, the degree of connectivity (ease of connection) for each characteristic value for each attribute between a number of strip members is defined based on past processing results for the strip members. Therefore, when a band-shaped member having a characteristic value that has not been processed before is included, there arises a problem that an optimum processing order cannot be determined for a large number of band-shaped members including the band-shaped member, and for each attribute. Since the degree of connection for each characteristic value is not necessarily defined from the viewpoint of quality, it cannot be determined whether or not the determined processing order of the band-shaped members is excellent in quality.

  The object of the present invention is to automatically consider processing order candidates in a processing line for a plurality of strip members even in the case where a plurality of strip members to be processed include a strip member for which no past processing record exists. It is an object of the present invention to provide a processing order creation device that can determine the processing order candidates and make the processing order candidates favorable from the viewpoint of quality.

  In order to solve the above-mentioned problem, a processing creation apparatus according to claim 1 of the present invention is a processing sequence for creating a processing sequence of strip members in a processing line that continuously processes a plurality of strip members with one or more processing facilities. A creation device for creating a list table in which a plurality of strip-shaped members to be processed are associated with one or more attributes of the strip-shaped member; and a connecting portion between the preceding strip-shaped member and the subsequent strip-shaped member An evaluation condition storage unit that stores a connection evaluation table in which an error level is set corresponding to a difference in attributes; a processing order candidate creation unit that creates a processing order candidate for a plurality of strip-shaped members included in the list table; Based on the connection evaluation table, for each connection portion of the preceding band member and the subsequent band member included in the processing order candidates, for each processing facility and for each attribute type The processing order evaluation means for obtaining the error level and summing the error levels to obtain a processing order error level; and the processing order candidate creation means for creating the processing order candidate, and the processing order evaluation means for performing the processing. A repetitive execution means for repeatedly executing the processing for obtaining the processing order error level for the order candidates for a preset number of times N, and selecting the processing solution with the smallest processing order error level among the N processing order candidates as the best solution And a best solution selection means.

  In the processing order creation device according to the first aspect, the iterative execution unit causes the processing order candidate creation unit to create a processing order candidate, and the processing order evaluation unit performs processing for obtaining a processing order error level for the processing order candidate in advance. The best solution selecting means selects the one with the smallest processing order error level as the best solution from the N processing order candidates. As a result, even when a plurality of strip-shaped members to be processed include a strip-shaped member for which no past processing record exists, the processing order candidates in the processing line can be automatically determined for the plurality of strip-shaped members, and the processing order candidates can be selected. Good quality can be achieved.

The processing order creation device according to claim 2 is the processing order creation device according to claim 1, wherein the processing order evaluation means further obtains a change rate of each of the connection sections and calculates a sum of the change rates. The best solution selecting means selects the best solution having the smallest sum of the change rates of the connecting portions when the processing order error level is the same.
Further, the processing sequence creation device according to claim 3 is the processing sequence creation device according to claim 2, wherein the processing sequence evaluation means calculates the total of the processing sequence error level and the change rate for each processing facility. Further, the processing order error level and change rate are corrected by a preset weighting coefficient.

  As described above, according to the processing sequence creation device of the present invention, even when a plurality of strip-shaped members to be processed include a strip-shaped member for which no past processing results exist, processing in a processing line for a plurality of strip-shaped members. The order candidates can be automatically determined, and the processing order candidates can be made favorable from the viewpoint of quality.

1 is a block diagram showing a schematic configuration of a computer network system in which processing order creation devices according to an embodiment of the present invention are constructed in a distributed manner. It is a block diagram which shows the structure of the process order preparation part shown by FIG. It is a flowchart for demonstrating the preparation method of the best solution by the process order preparation apparatus which concerns on embodiment of this invention.

Hereinafter, a steel sheet processing sequence creation apparatus according to an embodiment of the present invention will be described with reference to the drawings.
In addition, the steel plate processing order creation apparatus according to the present embodiment (hereinafter simply referred to as “processing order creation apparatus”) is applied to a continuous processing line for steel sheets in an iron factory. In such a continuous processing line of steel plates, for example, various processing facilities such as pickling equipment, annealing equipment, rolling equipment, plating equipment are arranged along the conveying path of the steel plates, and these processing equipment are A steel plate as a product material paid out from the coil is continuously processed to produce a steel plate (for example, a galvanized steel plate) as a product. In the continuous processing line for steel plates, from the viewpoint of manufacturing efficiency, the tip of the steel plate (following steel plate) to be processed next is connected to the rear end of the steel plate being processed by the processing equipment (preceding steel plate) by welding or the like. Therefore, it is desirable to continuously perform processing on as many steel plates as possible.

  FIG. 1 schematically shows a configuration example of a computer network system in which processing order creation apparatuses according to the present embodiment are constructed in a distributed manner. The computer network system 10 includes a host computer 12 for managing the entire steel factory, an operation terminal 14 connected to the host computer 12 via a network such as a LAN (Local Area Network), and a processing sequence creation server. 16 is provided.

  The operation terminal 14 has a function of inputting various data and commands to the host computer 12. Specifically, the operation terminal 14 functionally includes a processing target coil input unit 18, a processing order result confirmation unit 20, a processing order result reflection instruction unit 22, and a setting change instruction unit 24. The processing target coil input unit 18 inputs a planning target period (for example, one week), and processes data of the processing target coil to be processed in the continuous processing line within the planning target period from the actual product DB (database) 28. The date and the planning target period are compared, extracted, and output to the list creation unit 26 of the host computer 12. Further, the processing target coil input unit 18 can preset the order of a specific coil in the extracted data. The coil set by the operator is excluded from the calculation target. When the target data is determined in this way, the calculation target start is input from the processing target coil input unit 18.

The list creation unit 26 of the host computer 12 is configured as list creation means in the present invention. The list creation unit 26 expands the actual product data, and creates a list table L in which actual product numbers, standard codes, and one or more attributes of these steel plates are associated with each other.
Here, the attributes of the steel sheet include, for example, quantitative data expressed quantitatively such as the width and thickness of the steel sheet, and qualitative data that defines the necessity and processing method of a specific process for the steel sheet. ing. However, qualitative data can also be handled as quantitative data by giving grades (error levels) in stages according to the necessity of processing and the contents of processing methods. For this reason, in the present embodiment, the qualitative attribute of the steel plate will be described as being represented by quantitative data. That is, the present invention focuses on the fact that quantitative data and qualitative data can be handled without distinction by using graded grades (error levels).

  In the following [Table 1], an example of the list table L created by the list creation unit 26 is shown. In this list table L, an example in which three types of attributes (attributes 1-3) are set is shown, but the set attributes can be set to an arbitrary number as necessary. Here, attribute 1 is the sheet width of the steel sheet, attribute 2 is the sheet thickness of the steel sheet, and attribute 3 is the annealing temperature for the steel sheet.

  As shown in FIG. 1, the processing order creation server 16 includes a processing order creation unit 30 and a DB (database) 32, and the list table created by the list creation unit 26 is sent to the processing order creation unit 30. A connection evaluation table C, a weighting table A, and a weighting table B are stored in the DB 32 that is an evaluation condition storage unit.

  In the following [Table 2], an example of the connection evaluation table C stored in the DB 32 is shown. Table 3 below shows an example of the weighting table A stored in the DB 32.

  The weighting table A is an attribute (attribute type) to be considered for each of a plurality of processing facilities (in this embodiment, the pickling facility F1, the rolling facility F2, the annealing facility F3, and the plating facility F4) and should be considered. A weighting coefficient EC associated with the attribute type is set.

  The column (matrix) in which @ is written in the weighting table A indicates that the attribute type corresponding to the matrix is not an evaluation target to be considered. For example, in [Table 3], for the rolling equipment F2, only attribute 1 (sheet width) and attribute 2 (sheet thickness) are evaluated, and attribute 3 (annealing temperature) is not evaluated, Since the weighting coefficient EC of attribute 1 is 3 and the weighting coefficient EC of attribute 2 is 2, the weight (importance) of attribute 1 is higher than attribute 2 in determining the processing order candidate of the steel plate to be processed. Represents high. The connection evaluation table C is set for each attribute type to be considered according to the processing facility.

Here, [Table 2] shows an example of the connection evaluation table C set in association with the rolling equipment F2 and attribute 1 (sheet width). In this connection evaluation table C, an error level EL _A corresponding to a difference in attributes (attribute deviation or combination in the case of qualitative attributes) between the preceding steel sheet and the subsequent steel sheet is set. The column of the operation condition in the connection evaluation table represents an operational category such as the type of rolling roll in the rolling equipment F2 and the size of the rolling reduction with respect to the steel plate.

The connection evaluation table C in Table 2 shows a case where one attribute 1 (plate width) is evaluated using two evaluation criteria of a change rate CR and a change amount CD. Here, for example, assuming that the attribute value of the preceding steel plate is V1 and the attribute characteristic value of the subsequent steel plate is V2, the change rate CR of the attribute is calculated by the following equation (1).
Rate of change CR = (V2−V1) / V1 × 100 (%) (1)
The change amount of attribute 1 is (V2-V1). For example, in the case of a plate width, it is calculated by (w2-w1) (mm).
In the connection evaluation table, an error level EL _A corresponding to the attribute deviation quantified step by step is set. For example, in the operation condition C1, the error level with respect to the deviation of the attribute 1 is set in four stages.

  Next, the best solution selection process executed by the processing order creation unit 30 will be described with reference to the block diagrams of FIGS. 1 and 2 and the flowchart of FIG. As shown in FIG. 2, the processing order creation unit 30 includes a processing order candidate creation means 34, a processing order evaluation means 36, an iterative execution means 38, and a best solution selection means 40 in order to determine the best solution described later. Yes.

  In step S200 of FIG. 3, the processing order creation unit 30 sets “1” to the loop counter number n. In step S202, the processing order candidate creation unit 34 creates processing order candidates PR for a plurality of steel plates included in the list table L. The processing order candidate PR as the initial solution may be basically a plurality of steel plates included in the list table arranged in any order, but here, the initial solution is a plurality of steel plates to be processed. It is assumed that they are arranged in the order described in the list table L. When the loop counter number n is 2 or more, the processing order candidate creation means 34 replaces the order of one or more steel plates included in the processing order candidate PR created last time according to a predetermined rule. A processing order candidate PR different from the already created (n-1) processing order candidates PR is created.

The order of the steel plates may be changed by using a generally known method such as a local search method, a guidance search method, a genetic algorithm, a tabu search method, or an annealing method. That's fine.
In step S204, the processing order evaluation means 36 determines, based on the connection evaluation table C, the individual error level ELi for the connecting portion between the preceding steel plate and the subsequent steel plate included in the processing order candidate PR for each processing equipment. Each attribute type to be considered is obtained. Here, when the processing order candidate PR is composed of k steel plates, the processing order candidate PR includes (k−1) connection portions.

In step S206, the processing order evaluation unit 36 calculates the weighting error level ELw by correcting the individual error level ELi based on the weighting table A by the weighting coefficient EC set in advance for each processing equipment and for each attribute type. To do.
The calculation method of the individual error level ELi and the weighted error level ELw will be specifically described with reference to [Table 2] and [Table 3].

For example, if the operation condition is C1 and the change rate CR of attribute 1 of a certain connection portion is 8% and the change amount CD is 90 mm, which is included in the processing order candidate PR, this attribute 1 has two allowable ranges. (EL _A = 0 for CR and EL _A = 10 for CD), but of any applicable tolerance range, the tolerance level EL _A is high (EL1 = 10). (That is, take the AND condition of both conditions). In this case, as shown in [Table 3], since “3” is set as the weighting coefficient EC of attribute 1, the weighting error level ELw is EL1 × EC = 30. The allowable range shown in the lower part of the connection evaluation table C excludes the overlapping range with the allowable range shown in the upper part.

  In step S208 and step S210, the processing order evaluation means 36 calculates the change rate RA of each attribute for each processing facility and for each attribute type for (k−1) connection units included in the processing order candidate PR. And calculating the weighting rate of change RAw by correcting the rate of change RA with a predetermined weighting factor RC preset for each processing facility. The change rate RA is calculated by the following equation (2).

Rate of change RA = | V ₂ −V ₁ | / V ₁ × 100 (%) (2)
Here, the weighting coefficient RC is preset in the weighting table B, and this weighting table B is stored in the DB 32 together with the weighting table A and the connection evaluation table C. In the following [Table 4], an example of the weighting table B stored in the DB 32 is shown.

  The weighting table B in [Table 4] includes attributes 4-6 (attribute types) to be considered for each of a plurality of processing facilities (in this embodiment, pickling facility F1, rolling facility F2, annealing facility F3, and plating facility F4). ) And a weighting coefficient RC associated with the attribute type to be considered. Here, attribute 4 is a plate thickness change rate on the entrance side with respect to the processing facility, attribute 5 is a plate thickness change rate on the exit side with respect to the processing facility, and attribute 6 is a plate width change rate on the entrance side with respect to the processing facility. Note that the column (matrix) in which @ is written in the weighting table B indicates that, as with the weighting table A, the attribute type corresponding to the matrix is not an evaluation target.

  In step S212, the processing order evaluation means 36 sums the weighting error levels ELw for each processing equipment (F1 to F4) and each attribute calculated for the processing order candidate PR, and sets the processing order error level ELt of the processing order candidate Pr. calculate. In step S214, the processing order evaluation unit 36 calculates the change rate evaluation value ER by summing up the (k-1) weighted change rates RAw for each processing facility and each attribute type.

In step S216, the repeated execution means 38 determines whether or not the loop counter number n is 2 or more. At this time, if the loop counter number n is not 2 or more (n = 1), in step S218, the processing order candidate PR is stored in the internal memory 42 (see FIG. 1), and the processing order candidate PR. Are stored in the internal memory 42 as ELt ₀ and ER ₀ , respectively. In step S220, the repeated execution means 38 increments (+1) the loop counter number n and returns the processing routine to step S202.

Further, when the number of loop counter n is 2 or more in step S216, in step S222, the processing order evaluation means 36, already calculated this time and the order of processing error level ELT ₀ stored in the internal memory 42 The processing order error level ELt is compared.
At this time, if ELt is smaller than ELt ₀ , the processing order evaluation unit 36 stores the processing order candidate PR corresponding to ELt in the internal memory 42 in step S 224, and sets ELt as a new ELt _0. To store.

On the other hand, if ELt is a value equal to or greater than ELt ₀ , the processing order evaluation means 36, in step S226, the processing order error level ELt ₀ already stored in the internal memory 42 and the processing order error level calculated this time. It is determined whether ELt matches. At this time, if ELt ₀ and ELt do not match, the processing routine is shifted to step S232 in step S220.

If the processing order error level EL ₀ matches the processing order error level ELt, the processing order evaluation unit 36 determines that the change rate evaluation value ER ₀ already stored in the internal memory 42 is the same as the change rate evaluation value ER ₀ in step S228. The change rate evaluation value ER calculated this time is compared. When ER is smaller than ER ₀ , the processing order candidate PR corresponding to ER is stored in the internal memory 42 in step S230, and ER is set as a new ER _0. Store in the internal memory 42. When ER is equal to or greater than ER ₀ , the process routine is shifted to step S232.

  In step S232, the repeated execution means 38 determines whether or not the loop counter number n has reached a preset number of evaluations N. If the loop counter number n has not reached the evaluation number N, the repeat execution means 38 increments (+1) the loop counter number n in step S220 and returns the processing routine to step S202. If the loop counter number n has reached the number of evaluations N, the best solution selection means 40 selects the processing order candidate PR stored in the internal memory 42 as the best solution in step S234.

  The best solution selection means 40 outputs the processing order candidate PR selected as the best solution to the output information processing unit 46. When the processing order candidate PR is input from the processing order creation unit 30, the output information processing unit 46 arranges the steel plates (actual product numbers) included in the processing sequence candidate PR in the processing order and associates them with the actual product numbers. Then, at least a standard code, a weighted error level ELw and weighting change rate RAw for the connection part are arranged (in the order of insertion), and the order of insertion is displayed on the display information on the screen such as the CRT or on the form. Displayed as description information by the processing order result display unit 58.

  The operator of the operation terminal 14 determines that the content of the insertion order displayed by the processing order result display unit 58 of the processing order creation server 16 is not problematic according to the operation standard of the continuous processing line. A predetermined confirmation operation is performed on the processing order result confirmation unit 20. As a result, the processing order result reflection instruction unit 22 outputs a processing order candidate PR (best solution) confirmation command to the processing order result reflection unit 60 of the host computer 12. Upon receiving this confirmation command, the processing order result reflection unit 60 sets the processing order candidate PR as the processing order in which the processing order candidate PR has been determined, and this processing order is stored in the DB 28 as a part of the production plan data (for example, charging in the same operation unit of the processing equipment). (Sequential number).

  In addition, the operator of the operation terminal 14 performs a predetermined invalid operation on the processing order result confirmation unit 20 of the operation terminal 14 when the content of the weekly plan has some problem in light of the operation standard of the continuous processing line. Execute. Thereby, it becomes possible for the operator to perform editing work such as replacement and deletion of steel sheets to be processed within a predetermined period and editing (reviewing) work on Tables 1 to 4 by the setting change instruction unit. Then, the processing target coil edited by the setting change instruction unit is input again to the processing target coil input unit 18, and a new processing order candidate PR (secondary best solution) is created based on the data.

  In the processing order creation device of the present embodiment described above, the processing order evaluation means 36 obtains processing order error levels ELt for N processing order candidates PR, respectively, and the best solution selection means 40 has N processing order candidate PRs. By selecting the one with the smallest processing order error level ELt as the best solution, among the N processing order candidates PR created for the plurality of steel plates set in the list table L, the processing order error level ELt is The smallest one can be selected as the best solution.

  Here, for the processing order error level ELt, the attribute types to be considered for each processing facility stored in advance in the connection evaluation table C, and the allowable attribute differences of the connection portions quantified step by step for each attribute type Since it is calculated based on the weighting coefficient EC set in advance for each range and processing equipment, even when a plurality of steel plates to be processed include steel plates for which no past processing results exist, processing for a plurality of steel plates The forward candidate PR (best solution) can be automatically determined.

Further, in the processing sequence creation apparatus of this embodiment, when setting the attribute type for each processing facility, the allowable range of the attribute difference in the connection unit, and the weighting coefficient for each processing facility, there are quality requirements for each. Since it is set in consideration, the processing order candidate PR selected as the temporary best solution can be improved from the viewpoint of quality.
Although it is desirable to set the evaluation number N as large as possible, the load on the processing order creation server 16 increases rapidly as the evaluation number N increases. The upper limit value of the number of evaluations N is limited according to the 16 processing capabilities.

  Further, in the processing order creation device of this embodiment, when the processing order error levels ELt of a plurality of processing order candidates PR are the same, the best solution selection means 40 selects the processing order candidate PR having a small change rate evaluation value ER. Select as solution. As a result, among the processing order candidates PR having the same processing order error level ELt, the one with the smallest change rate evaluation value ER can be selected as the best solution. For example, a plurality of attribute values to be considered are completely the same. In the case where there are strip members, it is possible to determine a processing order candidate (best solution) in which those strip members are processed continuously.

  In addition, in the above embodiment, although the case where this invention was applied to the continuous processing line of a steel plate was demonstrated, this invention is not restricted to this, The strip | belt-shaped member which is a several process target is used between process targets. It can be widely applied to the creation of a production plan that is selected in consideration of the predetermined constraint conditions that should be satisfied in step (1) and determines the order in which the continuous processing line is input.

DESCRIPTION OF SYMBOLS 10 Computer network system 12 Host computer 14 Operation terminal 16 Processing order creation server 18 Processing target coil input part 20 Processing order result confirmation part 22 Processing order result reflection instruction part 24 Setting change instruction part 26 List creation part 28 DB (actual product data) )
30 processing order creation unit 34 processing order candidate creation means 36 processing order evaluation means 38 repetitive execution means 40 best solution selection means 42 internal memory 46 output information processing section 58 processing order result display section 60 processing order result reflection section

Claims (3)

A processing sequence creation device for creating a processing sequence of strip members in a processing line that continuously processes a plurality of strip members with one or more processing facilities,
A list creation means for creating a list table in which a plurality of strip members to be processed are associated with one or more types of attributes of the strip members;
An evaluation condition storage means in which a connection evaluation table in which an error level is set corresponding to a difference in attributes in the connection portion between the preceding belt-like member and the subsequent belt-like member is stored;
Processing order candidate creation means for creating processing order candidates for a plurality of strip-shaped members included in the list table;
Based on the connection evaluation table, the error level is determined for each of the processing facilities and for each of the attribute types for each connection portion of the preceding band member and the subsequent band member included in the processing order candidate, A processing order evaluation means for summing up error levels to obtain a processing order error level;
Repetitive execution means for causing the processing order candidate creation means to create the processing order candidates and for causing the processing order evaluation means to repeatedly execute a process for obtaining the processing order error level for the processing order candidates for a preset number of times N. ,
Best solution selection means for selecting, as the best solution, the one with the smallest processing order error level among the N processing order candidates;
A processing sequence creation device characterized by comprising: The processing order evaluation means further calculates a change rate of each of the connection parts, calculates a sum of the change rates,
2. The processing order creation according to claim 1, wherein, when the processing order error level is the same, the best solution selecting means selects the one having the smallest sum of the change rates of the connection portions as the best solution. apparatus.   The processing order evaluation means corrects the processing order error level and the rate of change by a preset weighting factor for each processing facility when calculating the sum of the processing order error level and the rate of change. The processing order creation device according to claim 2, wherein

Images