Method for analyzing the tendency of defects to occur

Abstract

To provide a defect occurrence tendency analysis method capable of sufficiently contributing to yield improvement of a sheet-like product by sufficiently investigating a cause of defect occurrence of the sheet-like product.SOLUTION: A method for analyzing a tendency related to defect occurrence on the basis of a defect that occurs in a plurality of sheet-like products S2 manufactured by cutting a long raw material S1, or the like includes: a step ST11 for marking a position information mark such that a position information mark M representing position information in a longer direction and in a width direction in the raw material exists in each of the plurality of sheet-like products to uncut raw material S1 or the like; a step ST12 for inspecting the sheet-like products; a reading step ST13 for acquiring the position information in the raw material of the sheet-like products by reading the position information mark marked on the sheet-like products; and a step ST14 for specifying the position information in the raw material of the defect existing in the sheet-like products, and analyzing the tendency related to the defect occurrence on the basis of the specified position information.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to a method for analyzing the tendency of defect occurrence, which is based on defects generated in a plurality of sheet-like products manufactured by feeding out, conveying, and cutting a long web wound in a roll shape, such as a polarizing film or a retardation film, or by feeding out, conveying, and cutting a long web wound in a roll shape to cut out a large sheet-like intermediate body and then cutting the intermediate body. In particular, the present invention relates to a method for analyzing the tendency of defect occurrence that can sufficiently investigate the cause when a defect occurs in a sheet-like product and can sufficiently contribute to improving the yield of the sheet-like product.

Background Art

[0002] Conventionally, sheet-like products such as polarizing films and retardation films are manufactured as a plurality of sheet-like products by feeding out, conveying, and cutting a long web wound in a roll shape, or by feeding out, conveying, and cutting a long web wound in a roll shape to cut out a large sheet-like intermediate body and then cutting the intermediate body. Then, an optical display device such as a liquid crystal display device is manufactured by bonding each manufactured sheet-like product to an optical display unit such as a liquid crystal panel.

[0003] Here, if it is known at which position of the web before cutting or the web before cutting out the intermediate body the sheet-like product is after being bonded to the optical display unit, it is considered that it is possible to some extent to investigate the cause when a defect occurs in the sheet-like product and to contribute to improving the yield of the sheet-like product. For this reason, for example, Patent Document 1 proposes a method of marking a mark representing position information on the web before cutting. Specifically, Patent Document 1 proposes a method for marking a roll of raw material (a roll of raw material in Patent Document 1) before it is cut into a sheet-like product (an optical film in Patent Document 1) with a mark (a roll information holding means in Patent Document 1) that represents the longitudinal position information (a roll information in Patent Document 1) on the roll of raw material for the sheet-like product (see, for example, paragraphs 0131 to 0133 of Patent Document 1). According to the method described in Patent Document 1, it is possible to some extent to investigate the cause of abnormalities in sheet-like products and to improve the yield of sheet-like products.

[0004] However, the method described in Patent Document 1 has the problem that the positional information represented by the marks marked on the sheet-like product is only the positional information in the longitudinal direction on the raw material (paragraph 0131 of Patent Document 1), which makes it difficult to sufficiently investigate the cause of defects and, consequently, not to sufficiently contribute to improving the yield of the sheet-like product. Furthermore, even if the positional information represented by the marks on a sheet-like product includes not only positional information in the longitudinal direction of the raw material but also positional information in the width direction of the raw material, no conventional proposals have been made on how to utilize this information to help investigate the cause of defects. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2009-294645 [Overview of the project] [Problems that the invention aims to solve]

[0006] The present invention was made to solve the problems of the prior art described above, and aims to provide a method for analyzing the occurrence trend of defects that can thoroughly investigate the cause when defects occur in sheet-like products and can contribute significantly to improving the yield of sheet-like products. [Means for solving the problem]

[0007] To solve the above problem, the present invention provides a defect occurrence trend analysis method for analyzing the tendency related to defect occurrence based on defects that occur in multiple sheet-like products manufactured by unwinding, transporting, and cutting a long roll of raw material, or by unwinding, transporting, cutting out a large sheet-like intermediate, and then cutting the intermediate, wherein position information marks, which are marks representing position information in the longitudinal and width directions of the raw material, are present on each of the multiple sheet-like products before cutting or before cutting out the intermediate, A position information marking step; a sheet product inspection step for inspecting a plurality of the sheet products; a reading step for obtaining the position information of the sheet products in the raw material by reading the position information marks marked on the sheet products in which defects were detected in the sheet product inspection step; and an analysis step for identifying the position information of defects present in the sheet products in the raw material based on the position information of the sheet products in the raw material obtained in the reading step, and analyzing the trends related to the occurrence of defects based on the identified position information of defects present in the sheet products in the raw material. The analysis process includes: a raw material inspection step for inspecting the raw material before cutting or before cutting the intermediate product; a linking step for storing the location information of defects present in the raw material before cutting or before cutting the intermediate product, the type of defects present in the raw material, and the location where the raw material inspection step in which the defects were detected was performed, linked to the location information mark; the analysis step determines whether the location information of the defects present in the raw material in the identified sheet-like product matches the location information of the defects present in the raw material stored linked to the location information mark read in the reading step; and if they match, the analysis step stores the type of defects present in the raw material and the location where the defects were detected in the raw material, linked to the location information mark read in the reading step. A combination identification procedure for identifying a combination with an execution location; an extraction condition determination procedure for determining extraction conditions based on the combination of type and execution location identified in the combination identification procedure; a map image generation procedure for generating a map image which is an image plotting the locations of defects on the raw material based on the location information on the raw material of defects that conform to the extraction conditions determined in the extraction condition determination procedure, among all defects present in the raw material detected in the raw material inspection process; and a matching image generation procedure for generating a matching image which is an image representing the trend related to defect occurrence, by pattern matching the map image with a pre-prepared template image, in which only the locations of defects that match the template image among the locations of defects plotted on the map image are plotted. This invention provides a method for analyzing the tendency of defects to occur.

[0008] According to the present invention, in the position information marking process, position information marks, which are marks representing the position information in the longitudinal and width directions on the raw material, are marked on multiple sheet-like products such that each sheet-like product has a position information mark. Therefore, in the reading process, by reading the position information marks marked on each sheet-like product, the position information (position information in the longitudinal and width directions) of each sheet-like product on the raw material can be obtained. In the analysis process, based on the positional information of each sheet product in the raw material obtained in the reading process, the positional information of defects present in each sheet product detected in the sheet product inspection process can be identified. For example, if the positional information of defects present in each sheet product relative to each sheet product is unknown, the center position of each sheet product can be identified as the positional information of defects present in each sheet product in the raw material. Also, if the positional information of defects present in each sheet product relative to each sheet product is known, the positional information of defects present in each sheet product in the raw material can be identified based on the positional information of each sheet product in the raw material obtained in the reading process and the positional information of defects present in each sheet product relative to each sheet product. Therefore, in the analysis process, it is possible to analyze the trends related to the occurrence of defects based on the positional information of defects present in the sheet product in the raw material (positional information in the longitudinal and width directions) that has been identified. Therefore, if defects occur in sheet-like products, the cause can be thoroughly investigated, which can significantly contribute to improving the yield of sheet-like products.

[0009] The "positional information in the longitudinal and width directions of the raw material" represented by the positional information mark of the present invention means information including the position in at least the longitudinal direction (conveying direction) and the width direction perpendicular thereto of the raw material. The position in the longitudinal direction of the raw material may be represented by the distance from the leading edge (downstream end in the conveying direction) of the raw material, or by a number assigned sequentially according to the distance from the leading edge (conveying direction) of the raw material. The positional information mark may also represent information about the raw material for identifying multiple raw materials from each other. In addition, the positional information mark may also represent other incidental information. Furthermore, in this invention, the location information mark may be marked using ordinary colored ink or transparent ink, or it may be marked by laser engraving. Transparent ink is an ink that is invisible to the human eye under normal lighting, but becomes visible when irradiated with light of a specific wavelength and fluoresces. An example of transparent ink is UV ink that fluoresces when irradiated with ultraviolet light. Marking the location information mark with transparent ink has the advantage that the location information mark is invisible under normal lighting, and does not impair the appearance of the sheet product. An example of a location information mark is a one-dimensional code (barcode) or a two-dimensional code. An example of a two-dimensional code is DataMatrix® or QR code®. If the sheet product has a protective film or other layer that is removed when used (i.e., a layer that is not affected when the sheet product is used), it is preferable to mark the location information mark on that layer. Furthermore, the "tendency related to the occurrence of defects" in this invention refers to a wide range of tendencies related to the occurrence of defects, including not only the tendency for defects to occur (occurrence patterns) themselves, but also the tendency for defects to be detected in sheet products but difficult to detect in raw materials, the tendency of manufacturing process history information for raw materials in which defects are easily detected in sheet products, the tendency of sheet product varieties (raw material varieties) in which defects are easily detected, and the tendency of inspection methods and inspection conditions in which defects are easily detected when inspecting raw materials.

[0011] This invention In this context, "generating a map image" is not limited to the state in which the map image is actually displayed on a display device such as a monitor, but also includes the concept of identifying the location of defects in the raw material (location in the longitudinal and width directions of the raw material) and making it possible to display a map image. 。 As a "template image," you can prepare an image that reflects the pattern you want to extract from the map image, such as an image in which pixels are plotted at a constant period in the longitudinal direction according to the period (perimeter length) of the nip roller that transports the raw material.

[0013] This inventionIn this context, the raw material inspection process, which involves inspecting the raw material, is not limited to being performed at a single location in the raw material manufacturing process, but can also be performed at multiple locations. This invention According to the documentation, in the linking process, the location information of defects present in the raw material detected in the raw material inspection process, the type of defect present in the raw material, and the location where the raw material inspection process in which the defect was detected was performed are linked and stored in relation to a location information mark. Specifically, for example, the location information is linked and stored in relation to a location information mark marked on the same sheet product as the sheet product in which the defect detected in the raw material inspection process is located. Then, in the combination identification procedure of the analysis process, it is determined whether the location information of defects in the sheet product identified as described above (i.e., the location information of defects in the sheet product identified based on the location information of the sheet product in the sheet product acquired in the reading process, and hereafter referred to as "location information A") matches the location information of defects in the sheet product stored in association with the location information marks read in the reading process (i.e., the location information of defects in the sheet product detected in the sheet inspection process, and hereafter referred to as "location information B") matches. Here, the concept that location information A and location information B match is not limited to cases where the positions in the longitudinal and width directions of the sheet product included in location information A and the positions in the longitudinal and width directions of the sheet product included in location information B are exactly the same, but also includes cases where the positions in the longitudinal and width directions of the sheet product included in location information B are in a predetermined neighborhood area of ​​the positions in the longitudinal and width directions of the sheet product included in location information A. Then, in the combination identification procedure, if location information A and location information B match (in other words, if a defect in the raw material detected in the raw material inspection process exists in the same location as a defect detected in the sheet product, i.e., if it is considered that the same defect was detected in both the sheet product and the raw material), the combination of the type of defect present in the raw material, which is stored in association with the location information mark read in the reading process, and the location where the raw material inspection process in which the defect in the raw material was detected was identified. If there are multiple location information Bs (i.e., there are multiple location information for defects present in the raw material, which are stored in association with the location information marks read in the reading process), then, for example, it is sufficient to determine whether the location information B of the defect with the largest area matches location information A. In the extraction condition determination procedure of the analysis process, the extraction conditions are determined based on the combination of type and execution location identified in the combination identification procedure. Then, in the map image generation procedure, a map image is generated in which the locations of defects on the raw material are plotted based on the location information of defects on the raw material that meet the extraction conditions determined in the extraction condition determination procedure, out of all defects present in the raw material detected in the raw material inspection process. In other words, This invention In this map image generation procedure, a map image is generated in which only the locations of defects in the raw material that meet the same conditions (combination of defect type and location of raw material inspection process) as the defects detected in the sheet product (defects detected in the corresponding raw material) are plotted. Therefore, for example, by visualizing the matching image generated from this map image in the matching image generation procedure, it is possible to grasp the trends related to the occurrence of defects that may be detected in sheet-like products (trends in the occurrence (occurrence patterns) of defects).

[0014] This invention In the case where a large number of sheet products are found to have defects in the sheet product inspection process, it is preferable, for example, that in the combination identification procedure, a plurality of combinations of type and execution location are identified for each of the plurality of sheet products in which defects were found in the sheet product inspection process, and that in the extraction condition determination procedure, a majority of the combinations of type and execution location identified in the combination identification procedure are determined as the extraction conditions. As described above, by determining a majority of the identified combinations of types and execution locations as extraction conditions, even if a large number of sheet products have defects detected in the sheet product inspection process, and a large number of defects are detected in the raw material inspection process, a map image and matching image will be generated with an appropriate number of defect locations plotted for understanding trends. This is expected to allow for an appropriate understanding of the trends related to the occurrence of defects most likely to be detected in sheet products. The definition of "a large number of sheet products in which defects were detected during the sheet product inspection process" should be determined considering the computational load of the analysis process, but examples include cases where there are 50 or more sheets, or 100 or more sheets. Furthermore, the unit period for determining whether there are a large number of sheet products (in other words, the unit period for determining the majority of combinations as extraction conditions) can be, for example, a day or a week.

[0015] on the other hand, This invention In the case where the number of sheet products in which defects are detected in the sheet product inspection process is small, for example, in the combination identification procedure, it is possible to identify multiple combinations of the type and the execution location for each of the multiple sheet products in which defects are detected in the sheet product inspection process, and in the extraction condition determination procedure, all of the multiple combinations of the type and the execution location identified in the combination identification procedure are determined as the extraction conditions. If the number of sheet products in which defects are detected during the sheet product inspection process is small, the number of combinations of multiple types and execution locations identified in the combination identification procedure is also expected to be small. Therefore, even if all of the identified combinations of multiple types and execution locations are determined as extraction conditions, as described above, a map image and matching image will be generated with an appropriate number of defect locations plotted, making it possible to appropriately grasp the trends related to the occurrence of defects that may be detected in sheet products.

[0016] Also, This invention In cases where defects detected in the sheet-like product inspection process are often not detected in the raw material inspection process (in other words, where defects detected in the sheet-like product inspection process are rarely detected in the raw material inspection process), it is possible to output this situation as a trend related to defect occurrence without generating map images or matching images. In other words, This inventionIn this case, for example, in the combination identification procedure, for each of the plurality of sheet-like products in which defects are detected in the sheet-like product inspection process, combinations of the plurality of types and execution locations are identified. When the number of combinations of the plurality of types and execution locations identified in the combination identification procedure is less than half of the total number of defects detected in the sheet-like product inspection process for the plurality of sheet-like products, the analysis process does not execute the extraction condition determination procedure, the map image generation procedure, and the matching image generation procedure, and outputs, as an index representing the tendency related to the occurrence of defects, that the number is less than half (that is, the number of defects detected in the sheet-like product inspection process is less likely to be detected in the raw material inspection process). It is preferable to have an index output procedure. As described above, without executing the extraction condition determination procedure, the map image generation procedure, and the matching image generation procedure, in the index output procedure, the number of combinations of the plurality of types and execution locations identified in the combination identification procedure (in other words, the number considered to have the same defect detected in both the sheet-like product and the raw material) is less than half of the total number of defects detected for the plurality of sheet-like products. By outputting this as an index representing the tendency related to the occurrence of defects, it is possible to accurately grasp that defects are detected in the sheet-like product but not in the raw material without generating unnecessary map images and matching images, and to contribute to the investigation of the cause.

[0017] [[ID=??]]Also, To solve the aforementioned problem, The present invention A defect occurrence trend analysis method for analyzing the tendency related to defect occurrence based on defects that occur in multiple sheet-like products manufactured by unwinding, transporting, and cutting a long roll of raw material, or by unwinding, transporting, cutting a large sheet-like intermediate, and then cutting the intermediate, wherein the method involves marking the raw material, the intermediate, or the multiple sheet-like products before cutting or before cutting the intermediate, such that a position information mark, which is a mark representing the position information in the longitudinal and width directions of the raw material, is present on each of the multiple sheet-like products. A process comprising: a king process; a sheet product inspection process for inspecting a plurality of the sheet products; a reading process for obtaining the position information of the sheet products on the raw material by reading the position information marks marked on the sheet products in which defects were detected in the sheet product inspection process; an analysis process for identifying the position information of defects present in the sheet products on the raw material based on the position information of the sheet products on the raw material obtained in the reading process, and analyzing the trends related to the occurrence of defects based on the identified position information of defects present in the sheet products on the raw material; A raw material inspection process for inspecting the raw material before cutting or before cutting out the intermediate body; the position information of the defects existing in the raw material before cutting or before cutting out the intermediate body, which are detected in the raw material inspection process; the types of defects existing in the raw material; the execution location of the raw material inspection process in which the defects existing in the raw material are detected; and the manufacturing process history information of the raw material are associated and stored with the position information mark. of It should be noted that there seems to be an error in the text where "また、" is translated as "??" in the above translation. It should be translated as "Also,".has, and the analysis step determines whether the position information of the defect existing in the original fabric of the defect existing in the specified sheet product matches the position information of the defect existing in the original fabric stored in association with the position information mark read in the reading step. If they match, the combination specifying procedure that specifies the combination of the type of defect existing in the original fabric, the location where the original fabric inspection step in which the defect existing in the original fabric was detected was performed, and the manufacturing process history information of the original fabric, and based on the combination of the type and the execution location specified in the combination specifying procedure, an extraction condition determination procedure for determining extraction conditions, and among all the defects existing in the original fabric detected in the original fabric inspection step, the manufacturing process history information of the original fabric in which the defects that meet the extraction conditions determined in the extraction condition determination procedure exist is output as an index representing the tendency related to the occurrence of defects, and has 、 Method for analyzing the tendency of defect occurrence It is also offered as such.

[0018] This invention According to, unlike the above-mentioned direction method, in the association step, in addition to the position information of the defect existing in the original fabric detected in the original fabric inspection step, the type of defect existing in the original fabric, and the location where the original fabric inspection step in which the defect existing in the original fabric was detected was performed, the manufacturing process history information of the original fabric is associated with the position information mark and stored. Also, unlike the above-mentioned direction method, in the combination specifying procedure of the analysis step, the combination of the type of defect existing in the original fabric, the location where the original fabric inspection step in which the defect existing in the original fabric was detected was performed, and the manufacturing process history information of the original fabric is specified. And in the extraction condition determination procedure, based on the specified combination of the type and the execution location, extraction conditions are determined, and in the index output procedure, among all the defects existing in the original fabric, the manufacturing process history information of the original fabric in which the defects that meet the extraction conditions exist is output as an index representing the tendency related to the occurrence of defects. Therefore This inventionAccording to this, it is expected that the trends in manufacturing process history information of raw materials in which defects may be detected in sheet-like products can be appropriately grasped.

[0019] This invention In the case where a large number of sheet products are found to have defects in the sheet product inspection process, it is preferable, for example, that in the combination identification procedure, a plurality of combinations of type and execution location are identified for each of the plurality of sheet products in which defects were found in the sheet product inspection process, and that in the extraction condition determination procedure, a majority of the combinations of type and execution location identified in the combination identification procedure are determined as the extraction conditions.

[0020] Also, To solve the aforementioned problem, This invention A defect occurrence trend analysis method for analyzing the tendency related to defect occurrence based on defects that occur in multiple sheet-like products manufactured by unwinding, transporting, and cutting a long roll of raw material, or by unwinding, transporting, cutting a large sheet-like intermediate, and then cutting the intermediate, wherein the method involves marking the raw material, the intermediate, or the multiple sheet-like products before cutting or before cutting the intermediate, such that a position information mark, which is a mark representing the position information in the longitudinal and width directions of the raw material, is present on each of the multiple sheet-like products. A process comprising: a king process; a sheet product inspection process for inspecting a plurality of the sheet products; a reading process for obtaining the position information of the sheet products on the raw material by reading the position information marks marked on the sheet products in which defects were detected in the sheet product inspection process; an analysis process for identifying the position information of defects present in the sheet products on the raw material based on the position information of the sheet products on the raw material obtained in the reading process, and analyzing the trends related to the occurrence of defects based on the identified position information of defects present in the sheet products on the raw material; A raw material inspection step for inspecting the raw material before cutting or before cutting the intermediate, a linking step for storing the location information of defects present in the raw material before cutting or before cutting the intermediate, the type of defects present in the raw material, the location where the raw material inspection step in which the defects were detected was performed, and the manufacturing process history information of the raw material, linked to the location information mark. ofThe analysis step includes determining whether the location information of the defect present in the identified sheet-like product on the raw material matches the location information of the defect present in the raw material stored in association with the location information mark read in the reading step, and if they match, a combination identification step to identify the combination of the type of defect present in the raw material stored in association with the location information mark read in the reading step, the location where the raw material inspection step in which the defect present in the raw material was detected was performed, and the manufacturing process history information of the raw material, and the type identified in the combination identification step and The system comprises an extraction condition determination procedure that determines extraction conditions based on the combination of execution locations, and a determination procedure that performs a determination using a learning model to determine the tendency related to the occurrence of defects, wherein in the determination procedure, the manufacturing process history information identified in the combination identification procedure of the raw materials containing defects that conform to the extraction conditions determined in the extraction condition determination procedure, among all defects present in the raw materials detected in the raw material inspection process, is input to the learning model, and a determination result of whether or not the defects present in the raw materials conforming to the extraction conditions are detected as defects in the sheet-like product is output from the learning model. Missing Method for analyzing the tendency of point occurrence It is also offered as such.

[0021] This invention According to the procedure, in the determination process, the identified manufacturing process history information of raw materials containing defects that meet the extraction criteria is input into the learning model, and the learning model outputs a determination result indicating whether or not the defects present in the raw materials that meet the extraction criteria are detected as defects in the sheet-like product. therefore, This invention According to this, the likelihood of defects being detected (occurring) in sheet-like products can be evaluated (determined) based on the manufacturing process history information of the raw material. In addition, This invention In this context, the learning model may be generated by supervised learning using known input-output combinations as training data, or it may be generated by unsupervised learning. [Effects of the Invention]

[0022] According to the present invention, it is possible to thoroughly investigate the cause of defects when defects occur in sheet-like products, and this can greatly contribute to improving the yield of sheet-like products. [Brief explanation of the drawing]

[0023] [Figure 1] This figure schematically shows the general configuration of a system for performing an analysis method according to one embodiment of the present invention. [Figure 2] This is a flowchart showing the general steps of a manufacturing method for a sheet-like product to which the analysis method according to one embodiment of the present invention is applied. [Figure 3] This figure schematically illustrates the state of the raw material S1 and the sheet-like product S2 in the manufacturing method shown in Figure 2. [Figure 4] This is a flowchart showing the general steps of the analysis method according to the first embodiment. [Figure 5] This diagram schematically illustrates examples of map images and template images. [Figure 6] This is a flowchart showing the general steps of the analysis method according to the second embodiment. [Figure 7] This figure schematically illustrates the contents of analysis step ST26 of the analysis method according to the second embodiment. [Figure 8] This is a flowchart showing the general steps of the analysis method according to the third embodiment. [Figure 9] This is a flowchart showing the general steps of the analysis method according to the fourth embodiment. [Figure 10] This is a flowchart showing the general steps of the analysis method according to the fifth embodiment. [Figure 11] This figure shows a schematic configuration of a manufacturing apparatus for performing the analysis method according to a modified version of the present invention. [Modes for carrying out the invention]

[0024] Hereinafter, with reference to the attached drawings as appropriate, the defect occurrence tendency analysis method (hereinafter, as may be referred to simply as the "analysis method") according to embodiments of the present invention (first to fifth embodiments) will be explained using the case in which positional information marks are marked on the raw material before cutting as an example. Examples of sheet-like products to which the analysis method according to this embodiment can be applied include optical films such as polarizing films, phase difference films, visual compensation films, brightness enhancement films, and laminated films made by laminating two or more of these films. In this embodiment, the case in which the sheet-like product is a laminated film in which a phase difference film or the like is laminated on a polarizing film will be described as an example. Since the composition of each film is publicly known, a detailed explanation will be omitted here.

[0025] <System Configuration> Figure 1 is a schematic diagram showing the general configuration of a system for performing the analysis method according to this embodiment. In Figure 1, arrow X represents the transport direction (horizontal direction) of the raw material S1 for manufacturing the sheet-like product S2, arrow Y represents the width direction of the raw material S1 (horizontal direction perpendicular to the transport direction), and arrow Z represents the normal direction (vertical direction) of the surface of the raw material S1. As shown in Figure 1, the system 100 of this embodiment includes a sheet product manufacturing apparatus 100a located in a sheet product manufacturing process for manufacturing sheet product S2, and a reading device 6 located in a sheet product inspection process for inspecting sheet product S2.

[0026] The manufacturing apparatus 100a of this embodiment comprises an inspection device 1, a marking device 2, a cutting device 3, and a control and analysis device 4. The control and analysis device 4 is electrically connected to the inspection device 1, the marking device 2, and the cutting device 3. The manufacturing apparatus 100a of this embodiment also comprises a feed roller R1, a nip roller R2, and a conveyor R3. The manufacturing apparatus 100a of this embodiment is an apparatus that feeds out a long roll of raw material S1 wound in a roll shape onto a feed roller R1, transports it in the X direction by a nip roller R2, a conveyor R3, etc., and cuts it with a cutting device 3 to produce multiple sheet-like products S2. The raw material S1 is a long laminated film in which a phase difference film or the like is laminated onto a polarizing film, and is manufactured by known processes such as stretching, adhesive bonding, and lamination.

[0027] The inspection device 1 is a device that detects defects present in the raw material S1 by inspecting the raw material S1 before cutting. The inspection device 1 shown in Figure 1 comprises a light source 11 positioned on one side of the raw material S1 in the Z direction (below the raw material S1 in the example shown in Figure 1) and emitting light toward the raw material S1; an imaging means 12 positioned on the other side of the raw material S1 in the Z direction (above the raw material S1 in the example shown in Figure 1) and generating a transmitted image by receiving the light that has passed through the raw material S1 and forming an image (taking a picture); and an image processing means 13 that detects defects by applying known image processing, such as binarization, to the transmitted image input from the imaging means 12 to extract pixel regions with different brightness values ​​from other pixel regions. In this embodiment, the control and analysis device 4 also functions as the image processing means 13 of the inspection device 1, but it is also possible to provide an image processing means 13 separately from the control and analysis device 4.

[0028] As described above, the inspection device 1 is not limited to a configuration that detects defects based on a transmitted image. It is also possible to adopt a configuration that detects defects based on a reflected image generated by arranging a light source and an imaging means on one side in the Z direction relative to the raw material S1. Furthermore, if the raw material S1 (sheet-like product S2) is a polarizing film, the inspection device 1 can be configured to either place a light source and an inspection polarizing filter on one side of the raw material S1 in the Z direction and an imaging means on the other side of the raw material S1 in the Z direction, or to place a light source on one side of the raw material S1 in the Z direction and an inspection polarizing filter and imaging means on the other side of the raw material S1 in the Z direction, thereby detecting defects based on the crossed nicol image generated. Furthermore, the inspection device 1 can also employ a configuration that detects defects by combining two or more images from among transmitted images, reflected images, and crossed-nicol images. Figure 1 illustrates, for convenience, a case in which the raw material S1 is inspected at a single location using one inspection device 1. However, the present invention is not limited to this, and it is also possible to place inspection devices 1 in processes prior to the manufacturing apparatus 100a to inspect the raw material S1 at multiple locations.

[0029] The control and analysis device 4 (image processing means 13), which constitutes part of the inspection device 1, is capable of recognizing the location (XY coordinates) of defects in the transmission image. Therefore, for example, the control and analysis device 4 can recognize the location of defects in the raw material S1 based on the recognized location (XY coordinates) of defects in the transmission image, the separation distance L1 in the X direction between the feed roller R1 and the inspection device 1, and the amount of raw material S1 being transported measured by an encoder (not shown) attached to the nip roller R2.

[0030] The marking device 2 is a device that marks position information marks, which represent the position information on the raw material roll S1, onto the raw material roll S1 before cutting. Specifically, the control and analysis device 4 controls the marking device 2 so that multiple position information marks are marked on each of the multiple sheet-like products S2 after cutting. The marking device 2 of this embodiment is an inkjet-type marking device that uses transparent ink (specifically, UV ink), and is configured to mark positional information marks by ejecting transparent ink from a number of nozzles arranged along the Y direction. However, the marking device 2 can also be configured to use an inkjet method with ordinary colored ink or a laser engraving method. Furthermore, when using ink, it is possible to use a pen method for marking, not just an inkjet method. Since the specific configurations of these marking devices are publicly known, a detailed explanation is omitted here.

[0031] The cutting device 3 is a device that manufactures multiple sheet-like products S2 by cutting the raw material S1 using known processing methods such as punching or laser processing. Specifically, the control and analysis device 4 controls the cutting device 3 so that the raw material S1 is cut along predetermined cutting lines.

[0032] The control and analysis device 4 consists of a computer on which programs for controlling the operation of the marking device 2 and the cutting device 3, and programs for executing the analysis method described later, are installed.

[0033] The manufacturing apparatus 100a may also include a marking device (not shown) that marks the locations of defects detected by inspecting the raw material S1 with the inspection device 1, etc., before cutting. Specifically, the control and analysis device 4 may recognize the locations of defects detected by inspecting the raw material S1 with the inspection device 1, etc., and the control and analysis device 4 may control the marking device to mark the locations of defects. The marking device can be configured to use an inkjet method with ordinary colored ink, similar to the marking device 2, or to use a laser engraving method.

[0034] [Manufacturing method for sheet-like products] The following describes a method for manufacturing a sheet-like product S2 using a manufacturing apparatus 100a having the above configuration. Figure 2 is a flowchart showing the schematic steps of a manufacturing method for a sheet-like product to which the analysis method according to this embodiment is applied. As shown in Figure 2, the manufacturing method of this embodiment includes a position information marking step ST1, a cutting step ST2, and a recovery step ST3. Note that the position information marking step ST1 described here is the same as the position information marking steps ST11, ST22, ST32, ST42, and ST52 of the first to fifth embodiments described later. Figure 3 schematically illustrates the state of the raw material S1 and the sheet-like product S2 in the manufacturing method shown in Figure 2. Figure 3(a) schematically shows the state of the raw material S1 before the position information marking process ST1 is executed. Figure 3(b) schematically shows the state of the raw material S1 immediately after the position information marking process ST1 is executed. Figure 3(c) schematically shows the state of the raw material S1 (sheet-like product S2) immediately after the cutting process ST2 is executed. Figure 3(d) schematically shows the state of the sheet-like product S2 during the recovery process ST3 is executed. The following describes each process, ST1 to ST3, in order.

[0035] [Location information marking process ST1] As shown in Figure 3(a), it is assumed that defects F (in Figure 3(a), for convenience, all defects F are shown as black circles) are detected by inspecting the raw material S1 before cutting with the inspection device 1. As shown in Figure 3(b), in the position information marking process ST1, a position information mark M representing the position information of the raw material S1 is marked on the raw material S1 before cutting with the marking device 2. The position information represented by the position information mark M means information including at least the position (i.e., XY coordinates) of the raw material S1 in the longitudinal direction (conveying direction, X direction) and the width direction (Y direction). The position in the longitudinal direction of the raw material S1 may be represented by the distance from the leading edge of the raw material S1 in the longitudinal direction (conveying direction), or by a number assigned sequentially according to the distance from the leading edge of the raw material S1 in the longitudinal direction (conveying direction). The position information represented by the position information mark M may include, in addition to the position of the raw material roll S1 in the transport direction, information about the raw material roll S1 for identifying multiple raw material rolls S1 from each other, or other incidental information. The location information mark M in this embodiment is DataMatrix®, a type of two-dimensional code. However, the present invention is not limited to this, and various forms of marks can be used as the location information mark M, in addition to other two-dimensional codes such as QR code® and one-dimensional codes (barcodes), as long as they can represent the location information on the raw material S1.

[0036] In the position information marking process ST1, the marking device 2 marks the raw material S1 before cutting so that a position information mark M exists on each of the multiple sheet-like products S2 after cutting. Specifically, the planned cutting line CL, shown as a dashed line in Figure 3(b), is predetermined according to the dimensions and shape of the sheet-like product S2 and is stored in the control and analysis device 4. The planned cutting line CL is not actually drawn on the raw material S1, but is stored as an XY coordinate system based on, for example, the leading edge in the longitudinal direction of the raw material S1. In the example shown in Figure 3(b), the planned cutting line CL is grid-like, and the parts of the raw material S1 located within each rectangle (18 rectangles are shown in Figure 3(b)) demarcated by the planned cutting line CL become the sheet-like product S2 after cutting. Therefore, in the position information marking process ST1, the marking device 2 marks the position information mark M so that it exists within each rectangle demarcated by the planned cutting line CL, and so that it does not overlap with the planned cutting line CL. In the example shown in Figure 3(b), one position information mark M is marked within each rectangle, and the position of the position information mark M relative to the planned cutting line CL is set to be the same for all rectangles (at the center of each rectangle). However, the present invention is not limited to this. For example, when marking defect marks at the locations where defects exist, multiple position information marks M may be marked within each rectangle to reduce the possibility that the position information mark and the defect mark will overlap and make it impossible to read the position information mark. Also, when using multiple sheet-like products S2 stacked together, the positions of the position information marks M marked within each rectangle may be different from each other to reduce the risk of dents occurring in the overlapping parts due to the overlapping of position information marks on each sheet-like product S2.

[0037] When executing the position information marking process ST1, for example, the control analysis device 4 can calculate the timing at which a predetermined part of the raw material S1 reaches the marking device 2 based on the separation distance L2 in the X direction between the feed roller R1 and the marking device 2, and the amount of raw material S1 being transported measured by an encoder (not shown) attached to the nip roller R2. Then, for example, the control analysis device 4 controls the marking device 2 so that transparent ink is sprayed from the nozzle of the marking device 2 corresponding to the Y coordinate of the predetermined XY coordinate when a part having a predetermined XY coordinate with respect to the planned cutting line CL (a part to be marked with a position information mark M within each rectangle) reaches the marking device 2. As a result, a position information mark M is marked within each rectangle.

[0038] [Cutting process ST2] As shown in Figure 3(c), in the cutting process ST2, the raw material S1 marked with position information marks M is cut by the cutting device 3 along the planned cutting line CL (see Figure 3(b)) to produce multiple sheet-like products S2. When a die-cutting device is used as the cutting device 3, the transport of the raw material S1 is temporarily stopped when the planned cutting line CL of the raw material S1 reaches the cutting device 3, and the raw material S1 is cut. When a laser processing device is used as the cutting device 3, it is possible to cut the raw material S1 while it is being transported without stopping its transport. When executing the cutting process ST2, for example, the control analysis device 4 can calculate the timing at which the planned cutting line CL of the raw material S1, stored in the control analysis device 4, reaches the cutting device 3, based on the separation distance L3 in the X direction between the feed roller R1 and the cutting device 3, and the amount of raw material S1 being transported, measured by an encoder (not shown) attached to the nip roller R2. The control analysis device 4 then controls and drives the cutting device 3 at the timing when the planned cutting line of the raw material S1 reaches the cutting device 3. As a result, the raw material S1 is cut along the planned cutting line CL, and multiple sheet-like products S2 are manufactured.

[0039] [Recovery process ST3] As shown in Figure 3(d), in the recovery process ST3, the cut raw material S1 is transported by conveyor R3, and unwanted portions S11 (see Figure 3(c)) that do not become sheet-like products S2 are removed by a known removal device (not shown). Then, the sheet-like products S2 are transported by conveyor R3 and recovered by gravity dropping from the downstream end of the conveyor R3 in the transport direction.

[0040] As shown in Figure 1, the sheet-like product S2 manufactured as described above is transported to the sheet-like product inspection process and inspected visually by, for example, an inspector 5. However, this is not the only option, and inspection can also be performed using an optical automatic inspection device (not shown). Furthermore, for example, the sheet-like product manufacturing process may be carried out at a sheet-like product manufacturing plant, and the sheet-like product inspection process may be carried out at an optical display device manufacturing plant that manufactures optical display devices using the sheet-like product S2 shipped from the sheet-like product manufacturing plant. When inspecting the sheet-like product S2 at the optical display device manufacturing plant, although this is not the only option, known inspections such as lighting tests can be performed when the sheet-like product S2 is attached to an optical display unit. If a defect is detected during inspection, the position information mark M marked on the sheet-like product S2 is read by the reading device 6. The reading device 6 is a device that reads the position information mark M marked on the sheet-like product S2, and a known two-dimensional code reader or the like is used as the reading device 6.

[0041] By reading the position information mark M with the reading device 6, the position information (XY coordinates) of the sheet-like product S2 on the raw material S1 is acquired. The acquired position information is input to the control and analysis device 4. The acquired location information can be input to the control and analysis device 4 manually by a human, or the reading device 6 and the control and analysis device 4 can be electrically connected via a wireless communication line or the like so that the information is automatically input from the reading device 6 to the control and analysis device.

[0042] <Analysis method> The following describes the content of the analysis method according to this embodiment (first to fifth embodiment) using the system 100 having the configuration described above.

[0043] <First Embodiment> Figure 4 is a flowchart showing the schematic steps of the analysis method according to the first embodiment. As shown in Figure 4, the analysis method according to the first embodiment includes a position information marking step ST11, a sheet-like product inspection step ST12, a reading step ST13, and an analysis step ST14. The analysis method according to the first embodiment does not utilize the inspection results of the raw material S1. In other words, the inspection device 1 is not required to perform the analysis method according to the first embodiment. The following describes each process, ST12 to ST14, in order.

[0044] [Location information marking process ST11] In the position information marking process ST11, the marking device 2 is used to mark the raw material S1 before cutting with position information marks M, which are marks representing the position information in the longitudinal and width directions of the raw material S1, so that each of the multiple sheet-like products S2 has a position information mark M.

[0045] [Sheet-like product inspection process ST12] In the sheet-like product inspection process ST12, inspector 5 inspects multiple sheet-like products S2.

[0046] [Reading process ST13] In the reading process ST13, the reading device 6 is used to read the position information mark M that has been marked on the sheet-like product S2 in which defects were detected in the sheet-like product inspection process ST12, thereby obtaining the position information of the sheet-like product S2 on the raw material roll S1.

[0047] [Analysis process ST14] In the analysis step ST14, the position information acquired in the reading step ST13 is input to the control analysis device 4. Based on the position information of the sheet-like product S2 on the raw material S1 acquired in the reading step ST13, the control analysis device 4 identifies the position information of the defects present in the sheet-like product S2 detected in the sheet-like product inspection step ST12 on the raw material S1. Specifically, if the position information of the defects present in each sheet-like product S2 relative to each sheet-like product S2 is unknown (i.e., only the position information of each sheet-like product S2 on the raw material S1 is known), the control analysis device 4 identifies, for example, the center position of each sheet-like product S2 as the position information of the defects present in each sheet-like product S2 on the raw material S1. Furthermore, if the location information of defects present in each sheet-like product S2 is known, the control and analysis device 4 can identify the location information of the defects present in each sheet-like product S2 in the original roll S1 based on the location information of each sheet-like product S2 in the original roll S1 acquired in the reading process ST13 and the location information of defects present in each sheet-like product S2 in the original roll S1 based on each sheet-like product S2. Then, in the analysis step ST14, the control analysis device 4 is used to analyze the tendencies related to the occurrence of defects based on the positional information of the defects present in the identified sheet-like product S2 on the raw material S1.

[0048] Specifically, as shown in Figure 4, the analysis step ST14 of the analysis method according to the first embodiment includes a map image generation step ST141 and a matching image generation step ST142. Figure 5 schematically shows examples of map images and template images. Figures 5(a) to 5(d) show examples of map images, and Figures 5(e) to 5(i) show examples of template images.

[0049] (Map image generation procedure ST141) In the map image generation procedure ST141, the control and analysis device 4 generates a map image MP, which is an image plotting the locations of defects in the original roll S1 of the sheet-like product S2, based on the location information of the defects in the original roll S1 of the identified sheet-like product S2. In the map image MP shown in Figures 5(a) to 5(d), each area demarcated by a rectangle corresponds to the original roll S1, and the locations of defects in the original roll S1 of the sheet-like product S2 are indicated by black and white circles. In the actual map image MP, the pixels indicated by black and white circles are assumed to have different brightness values ​​(pixel values) from other pixels. In the example shown in Figures 5(a) to 5(d), the map image MP is generated by combining multiple continuously manufactured original rolls S1 (original rolls S1a to original rolls S1d) in the X direction, but this is not the only method, and a map image MP may be generated for each single original roll S1.

[0050] (Matching image generation procedure ST142) In the matching image generation procedure ST142, the control and analysis device 4 performs pattern matching between the map image MP and a pre-prepared template image TE as shown in Figures 5(e) to 5(i). This generates a matching image, which is an image representing the trend related to defect occurrence, in which only the defect locations that match the template image TE are plotted from the defect locations plotted on the map image MP. Specifically, the control and analysis device 4 scans the template image TE on the map image MP, extracts the pixel regions of the map image MP where the degree of agreement between the map image MP and the template image TE is above a predetermined threshold, and generates a matching image.

[0051] The template image TE shown in Figure 5(e) is an image plotting pixels indicated by white circles in the X direction with a constant period P, where this period P is set according to the period (outer circumference) of the feed roller R1, nip roller R2, etc. It is preferable to prepare multiple template images TE with different periods P according to the period of each roller used in the manufacturing process of the sheet-like product S2. In the actual template image TE, the pixels indicated by white circles have a different brightness value (pixel value) from other pixels, and have the same brightness value as the pixels indicated by black circles and white circles in the map image MP. The same applies to the template images shown in Figures 5(f) to 5(i). For example, by scanning the template image TE shown in Figure 5(e) in the X and Y directions on the map image MP shown in Figure 5(a) and performing pattern matching, the white circle pixels in the map image MP shown in Figure 5(a) are extracted, and a matching image is generated in which only these white circle pixels are plotted (black circle pixels are not plotted).

[0052] The three template images TE shown in Figure 5(f) are all images plotting pixels indicated by white circles that are densely packed in a predetermined shape. The template image TE shown at the top of Figure 5(f) is an example of an image plotting pixels that are densely packed in a parallelogram shape, the template image TE shown in the middle of Figure 5(f) is an example of an image plotting pixels that are densely packed in a linear shape, and the template image TE shown at the bottom of Figure 5(f) is an example of an image plotting pixels that are densely packed in a rhombus shape. However, these are not the only examples, and images plotting pixels that are densely packed in any shape, such as a triangle, square, or cross shape, can be used as template images TE. For example, by scanning the template image TE shown at the top of Figure 5(f) in the X and Y directions on the map image MP shown in Figure 5(b) and performing pattern matching, the white circle pixels in the map image MP shown in Figure 5(b) are extracted, and a matching image is generated in which only these white circle pixels are plotted (black circle pixels are not plotted).

[0053] The template image TE shown in Figure 5(g) is an image in which pixels indicated by white circles are plotted at specific Y coordinates (Y1) and specific X coordinates (X1). For example, by scanning the template image TE shown at the top of Figure 5(g) in the X direction on the map image MP shown in Figure 5(c) and performing pattern matching, the white circle pixels (pixels with a Y coordinate of Y1) in the map image MP shown in Figure 5(c) are extracted, and a matching image is generated in which only these white circle pixels are plotted (black circle pixels are not plotted).

[0054] The template image TE shown in Figure 5(h) is an image applied to the leading edge in the X direction of the map image MP (corresponding to the leading edge in the longitudinal direction of the raw material S1) and the trailing edge in the X direction of the map image MP (corresponding to the trailing edge in the longitudinal direction of the raw material S1). In other words, the template image TE shown in Figure 5(h) is an image used to extract defects present at the leading and trailing edges of the map image MP. For example, by scanning the template image TE shown in Figure 5(h) in the Y direction on the X-direction leading edge (the upper end of Figure 5(d)) of the map image MP shown in Figure 5(d) and performing pattern matching, the white circle pixels in the map image MP shown in Figure 5(d) are extracted, and a matching image is generated in which only these white circle pixels are plotted (black circle pixels are not plotted).

[0055] The template image TE shown in Figure 5(i) is an image in which pixels, indicated by white circles, are plotted so that they oscillate in the Y direction. While specific examples are omitted, by using this template image TE for pattern matching, it is possible to generate matching images that correspond to defects occurring at positions that oscillate in the width direction of the raw material S1 due to meandering or other factors in the raw material S1.

[0056] As described above, according to the analysis method of the first embodiment, in the map image generation procedure ST141 of the analysis step ST14, a map image MP is generated, which is an image in which the locations of defects present in the sheet-like product S2 on the raw material S1 are plotted. In the matching image generation procedure ST142, a matching image is generated, which is an image in which only the locations of defects that match the template image TE among the locations of defects plotted in the map image MP are plotted. Therefore, for example, by visually inspecting the matching image, it is possible to grasp the trend related to the occurrence of defects detected in the sheet-like product S2 (the trend of defect occurrence (occurrence pattern)).

[0057] <Second Embodiment> Figure 6 is a flowchart showing the schematic steps of the analysis method according to the second embodiment. As shown in Figure 6, the analysis method according to the second embodiment includes a raw material inspection step ST21, a position information marking step ST22, a linking step ST23, a sheet-like product inspection step ST24, a reading step ST25, and an analysis step ST26. Unlike the analysis method according to the first embodiment, the analysis method according to the second embodiment utilizes the inspection results of the raw material S1. For this reason, the analysis method according to the second embodiment has a raw material inspection step ST21 and a linking step ST23, which are not present in the analysis method according to the first embodiment. Furthermore, the content of the analysis step ST26 also differs from the analysis step ST14 of the analysis method according to the first embodiment. The following describes each process, ST21 to ST26, in order.

[0058] [Material inspection process ST21] In the raw material inspection process ST21, the raw material S1 before cutting is inspected using the inspection device 1 to detect defects present in the raw material S1. As mentioned above, the inspection device 1 can be placed in multiple locations, and the inspection device 1 can be configured to detect defects based on transmitted images, reflected images, or crossed-nicol images. Therefore, the control and analysis device 4 can determine the type of defect based on which image the defect was detected on, which location of the inspection device 1 it was detected on, and the location information of the defect on the raw material S1.

[0059] [Location information marking process ST22] In the position information marking step ST22, similar to the position information marking step ST11 of the analysis method according to the first embodiment, the marking device 2 is used to mark the raw material S1 before cutting with position information marks M, which are marks representing the position information in the longitudinal and width directions of the raw material S1, so that each of the multiple sheet-like products S2 has a position information mark M.

[0060] [Linking process ST23] In the linking process ST23, the control and analysis device 4 stores the location information of defects present in the raw material S1 before cutting, which were detected in the raw material inspection process ST21, the type of defects present in the raw material S1, and the location where the raw material inspection process ST21 in which the defects were detected (the location of the inspection device 1) was performed, by linking them to a location information mark M. Specifically, the control and analysis device 4 stores the location information mark M that is marked on the same sheet-like product S2 as the sheet-like product S2 where the defects detected in the raw material inspection process ST21 are located.

[0061] [Sheet-like product inspection process ST24] In the sheet-like product inspection process ST24, the inspector 5 inspects multiple sheet-like products S2, similar to the sheet-like product inspection process ST12 of the analysis method according to the first embodiment.

[0062] [Reading process ST25] In the reading step ST25, similar to the reading step ST13 of the analysis method according to the first embodiment, the reading device 6 is used to read the position information mark M marked on the sheet-like product S2 in which defects were detected in the sheet-like product inspection step ST24, thereby obtaining the position information of the sheet-like product S2 on the raw material S1.

[0063] [Analysis process ST26] In the analysis step ST26, similar to the analysis step ST14 of the analysis method according to the first embodiment, the position information acquired in the reading step ST25 is input to the control analysis device 4, and the control analysis device 4 identifies the position information of defects present in the sheet product S2 in the sheet product inspection step ST24 based on the position information of the sheet product S2 in the raw material S1 acquired in the reading step ST25. Then, in the analysis step ST26, similar to the analysis step ST14 of the analysis method according to the first embodiment, the control analysis device 4 is used to analyze the tendencies related to the occurrence of defects based on the positional information of the defects present in the identified sheet-like product S2 on the raw material S1. However, in the second embodiment, since the inspection results of the raw material S1 are used, the specific contents of the analysis process ST26 differ from those of the first embodiment, which does not use the inspection results of the raw material S1.

[0064] Specifically, as shown in Figure 6, the analysis step ST26 of the analysis method according to the second embodiment includes a combination identification step ST261, an extraction condition determination step ST262, a map image generation step ST263, and a matching image generation step ST264. Figure 7 is a diagram that schematically illustrates the contents of the analysis step ST26 of the analysis method according to the second embodiment.

[0065] (Combination identification procedure ST261) In the combination identification procedure ST261, the control analysis device 4 determines whether the location information of the defects present in the identified sheet-like product S2 on the raw material S1 (i.e., the location information of the defects present in the sheet-like product S2 on the raw material S1 identified based on the location information of the sheet-like product S2 on the raw material S1 acquired in the reading process ST25 (location information A)) matches the location information of the defects present in the raw material S1 stored in association with the location information mark M read in the reading process ST25 (i.e., the location information of the defects present in the raw material S1 detected in the raw material inspection process ST21 (location information B)). Specifically, if we consider the case where an image plotting the locations of all defects detected in the raw material inspection process ST21 on the raw material S1 (hereinafter referred to as the "raw image") is generated, we will consider the case where a raw image GI like the one shown in Figure 7(a) is generated. The black circles shown in Figure 7(a) are defects. Furthermore, in the example shown in Figure 7, we consider the case where five location information A are identified. In this case, as shown in Figure 7(b), the combination identification procedure ST261 determines whether the locations of the raw material S1 included in location information B in the X and Y directions are within a predetermined neighborhood area (five areas A1 to A5 shown by dashed lines in Figure 7(b)) of the locations of the raw material S1 included in location information A in the X and Y directions. In the example shown in Figure 7(b), black circles exist in regions A1, A2, A4, and A5, excluding region A3. This means that the positions of the raw material S1 contained in position information B in the X and Y directions are located in regions A1, A2, A4, and A5 (i.e., position information A and position information B corresponding to regions A1, A2, A4, and A5 match).

[0066] Then, in the combination identification procedure ST261, if the location information A and location information B match (in other words, if a defect in the raw material S1 detected in the raw material inspection process ST21 exists at the same location as the defect detected in the sheet-like product S2, that is, if it is considered that the same defect was detected in both the sheet-like product S2 and the raw material S1), the control analysis device 4 identifies the combination of the type of defect present in the raw material S1, which is stored in association with the location information mark M read in the reading process ST25, and the execution location of the raw material inspection process ST21 in which the defect present in the raw material S1 was detected. For example, consider a case where defects are detected in each of the 11 sheet-like products S2, numbered No. 1 to No. 11, and whether or not each defect was also detected in the raw material inspection process ST21 (whether or not location information A and location information B match), and if detected in the raw material inspection process ST21, the type of defect and the location where the raw material inspection process ST21 is performed are as shown in Table 1 below. [Table 1] In the case of Table 1 above, the combinations of the types of defects present in the raw material S1 and the locations where the defects in the raw material S1 were detected in the raw material inspection process ST21 (the nine hatched combinations in Table 1) corresponding to sheet-like products S2, excluding No. 3 and No. 9, for which no defects were detected in the raw material inspection process ST21, are identified in the combination identification procedure ST261.

[0067] (Extraction criteria determination procedure ST262) In the extraction condition determination procedure ST262, the control and analysis device 4 determines the extraction conditions based on the combination of type and execution location identified in the combination identification procedure ST261. Specifically, the control and analysis device 4 determines, for example, a majority of the combinations of types and execution locations identified in the combination identification procedure ST261 as extraction conditions. This method of determining a majority of combinations as extraction conditions is preferably used when there are many sheet-like products S2 in which defects are detected in the sheet-like product inspection process ST24. As shown in Table 1 above, of the nine identified combinations of type and execution location, five combinations, or more than half, have the defect type as "foreign matter" and the execution location as "stretching process." Therefore, the combination of "foreign matter" and "stretching process" is determined to be the extraction criterion.

[0068] For example, consider a case where defects are detected in two sheet-like products S2, No. 1 and No. 2, and whether or not each defect was also detected in the raw material inspection process ST21 (whether or not location information A and location information B match), and if it was detected in the raw material inspection process ST21, the type of defect and the location where the raw material inspection process ST21 was performed are as shown in Table 2 below. [Table 2] In the case of Table 2 above, the combinations of the types of defects present in the raw material S1 and the locations where the defects in the raw material S1 were detected in the raw material inspection process ST21 (two sets of hatched combinations in Table 2), corresponding to sheet-like product S2 (excluding No. 3, for which no defects were detected in the raw material inspection process ST21), are identified in the combination identification procedure ST261. Then, in the extraction condition determination procedure ST262, for example, all of the two sets of type and execution location combinations identified in the combination identification procedure ST261 can be used as extraction conditions. The method of determining all combinations as extraction conditions is preferably used when the number of sheet-like products S2 in which defects are detected in the sheet-like product inspection process ST24 is small. In the case shown in Table 2 above, the two identified combinations of types and execution locations, namely the combination where the defect type is "foreign matter" and the execution location is the "stretching process," and the combination where the defect type is "dent" and the execution location is the "stretching process," will be determined as extraction conditions.

[0069] In the extraction condition determination procedure ST262, the extraction conditions may be determined by fixing only one of the methods described above: the method of determining a majority of combinations as extraction conditions, or the method of determining all combinations as extraction conditions. Alternatively, it is possible to adopt a configuration in which both methods are switched depending on the number of sheet-like products S2 in which defects are detected in the sheet-like product inspection process ST24.

[0070] (Map image generation procedure ST263) In the map image generation procedure ST263, the control and analysis device 4 generates a map image MP, as shown in Figure 7(c), which plots the locations of defects in the raw material S1, based on the location information of defects in the raw material S1 that meet the extraction conditions determined in the extraction condition determination procedure ST262, out of all defects present in the raw material S1 detected in the raw material inspection process ST21 (black circles in Figures 7(a) and 7(b)). For example, in the case shown in Table 1 above, the combination of "foreign matter" and "stretching process" is determined as the extraction condition. Therefore, among all defects present in the raw material S1 detected in the raw material inspection process ST21, those defects whose type is "foreign matter" and whose location in the raw material inspection process ST21 where the defect was detected is the "stretching process" are selected, and a map image MP is generated in which the location of the selected defects in the raw material S1 is plotted.

[0071] (Matching image generation procedure ST264) In the matching image generation procedure ST264, the control and analysis device 4 performs pattern matching between the map image MP shown in Figure 7(c) and a pre-prepared template image TE (see Figures 5(e) to 5(i)). As shown in Figure 7(d), it generates a matching image MA, which is an image representing the trend related to defect occurrence. This matching image MA plots only the defect locations that match the template image TE among the defect locations plotted on the map image MP. The specific details of the matching image generation procedure ST264 are the same as those of the matching image generation procedure ST142 in the analysis method according to the first embodiment, so a detailed explanation is omitted here.

[0072] As described above, according to the analysis method of the second embodiment, in the map image generation procedure ST263 of the analysis step ST26, a map image MP is generated in which only the locations on the raw material S1 of defects that have the same conditions (combination of defect type and location of execution of raw material inspection step ST21) as the defects detected on the sheet-like product S2 (defects detected on the corresponding raw material S1) are plotted.Therefore, for example, by visualizing the matching image MA generated from this map image MP in the matching image generation procedure ST264, it is possible to grasp the trends related to the occurrence of defects that may be detected on the sheet-like product S2 (trends in defect occurrence (occurrence patterns)).

[0073] <Third Embodiment> Figure 8 is a flowchart showing the schematic steps of the analysis method according to the third embodiment. As shown in Figure 8, the analysis method according to the third embodiment includes a raw material inspection step ST31, a position information marking step ST32, a linking step ST33, a sheet-like product inspection step ST34, a reading step ST35, and an analysis step ST36. The analysis method according to the third embodiment, like the analysis method according to the second embodiment, is a method that utilizes the inspection results of the raw material S1. The contents of the raw material inspection step ST31, position information marking step ST32, linking step ST33, sheet-like product inspection step ST34, and reading step ST35 of the analysis method according to the third embodiment are the same as the contents of the raw material inspection step ST21, position information marking step ST22, linking step ST23, sheet-like product inspection step ST24, and reading step ST25 of the analysis method according to the second embodiment. For this reason, the following will mainly describe the part of the analysis step ST36 that differs from the analysis step ST26 of the analysis method according to the second embodiment, and will omit detailed explanations of the other steps.

[0074] [Analysis process ST36] In the analysis step ST36, similar to the analysis step ST26 of the analysis method according to the second embodiment, the position information acquired in the reading step ST35 is input to the control analysis device 4, and the control analysis device 4 identifies the position information of defects present in the sheet product S2 in the sheet product inspection step ST34 based on the position information of the sheet product S2 in the raw material S1 acquired in the reading step ST35. Then, in the analysis step ST36, similar to the analysis step ST26 of the analysis method according to the second embodiment, the control analysis device 4 is used to analyze the tendencies related to the occurrence of defects based on the positional information of the defects present in the identified sheet-like product S2 on the raw material S1. However, analysis step ST36 differs from analysis step ST26 in the following respects.

[0075] In the analysis step ST36, the control analysis device 4 determines whether the number of combinations of multiple types and execution locations identified in the combination identification procedure ST361, similar to the combination identification procedure ST261 of the analysis method according to the second embodiment, is less than half of the total number of defects detected in the sheet product inspection step ST34 for the multiple sheet products S2 (ST362 in Figure 8). Then, if the number of identified combinations of multiple types and execution locations is not less than a majority (i.e., a majority) (if "No" is displayed in ST362 in Figure 8), the control analysis device 4 executes the extraction condition determination procedure ST364, the map image generation procedure ST365, and the matching image generation procedure ST366. The contents of the extraction condition determination procedure ST364, the map image generation procedure ST365, and the matching image generation procedure ST366 are the same as the contents of the extraction condition determination procedure ST262, the map image generation procedure ST263, and the matching image generation procedure ST264 of the analysis method according to the second embodiment, respectively. On the other hand, if the number of identified combinations of types and execution locations is less than half (i.e., a majority) (if "Yes" is selected in ST362 in Figure 8), the control and analysis device 4 executes an index output procedure ST363, which outputs the fact that the number is less than half as an index representing the tendency related to the occurrence of defects, without executing the extraction condition determination procedure ST364, the map image generation procedure ST365, and the matching image generation procedure ST366.

[0076] For example, consider a case where defects are detected in each of the 11 sheet-like products S2, numbered No. 1 to No. 11, and whether or not each defect was also detected in the raw material inspection process ST31 (whether or not location information A and location information B match), and if detected in the raw material inspection process ST31, the type of defect and the location where the raw material inspection process ST31 is performed are as shown in Table 3 below. [Table 3] In the case of Table 3 above, the combinations of the types of defects present in the raw material S1 and the execution locations of the raw material inspection process ST31 in which defects present in the raw material S1 were detected (5 sets of hatched combinations in Table 3) corresponding to sheet products S2, excluding No. 2, No. 3, No. 5, No. 7, No. 9, and No. 11, in which no defects were detected in the raw material inspection process ST31, are identified in the combination identification procedure ST361. In other words, the number of combinations of types and execution locations identified in the combination identification procedure ST361 (5 sets) is less than half of the total number of defects (11) detected in the sheet product inspection process ST34 for multiple sheet products S2. Therefore, the control and analysis device 4 executes the index output procedure ST363 without performing the extraction condition determination procedure ST364, the map image generation procedure ST365, and the matching image generation procedure ST366, and outputs an index that is less than half as an indicator representing the tendency related to the occurrence of defects.

[0077] As explained above, according to the analysis method of the third embodiment, when certain conditions are met (when "Yes" is shown in ST362 in Figure 8), the index output procedure ST363 outputs an index representing the tendency related to defect occurrence, which is that the number of combinations of multiple types and execution locations identified in the combination identification procedure ST361 (in other words, the number of cases in which the same defect is considered to have been detected in both the sheet product S2 and the raw material S1) is less than half of the total number of defects detected for multiple sheet products S2. Therefore, without unnecessarily generating map images or matching images, it is possible to accurately grasp that defects are detected in the sheet product S2 but not in the raw material S1, and to contribute to investigating the cause.

[0078] <Fourth Embodiment> Figure 9 is a flowchart showing the schematic steps of the analysis method according to the fourth embodiment. As shown in Figure 9, the analysis method according to the fourth embodiment includes a raw material inspection step ST41, a position information marking step ST42, a linking step ST43, a sheet-like product inspection step ST44, a reading step ST45, and an analysis step ST46. The analysis method according to the fourth embodiment, like the analysis method according to the second embodiment, is a method that utilizes the inspection results of the raw material S1. The contents of the raw material inspection step ST41, position information marking step ST42, sheet-like product inspection step ST44, and reading step ST45 of the analysis method according to the fourth embodiment are the same as the contents of the raw material inspection step ST21, position information marking step ST22, sheet-like product inspection step ST24, and reading step ST25 of the analysis method according to the second embodiment. For this reason, the following will mainly describe the differences between the linking step ST43 and the analysis step ST46 of the analysis method according to the second embodiment, and will omit detailed descriptions of other steps.

[0079] [Linking process ST43] In the linking process ST43, the control and analysis device 4 links and stores the location information of defects present in the raw material S1 before cutting, which were detected in the raw material inspection process ST41, the type of defects present in the raw material S1, the execution location of the raw material inspection process ST41 in which the defects present in the raw material S1 were detected (the location of the inspection device 1), and the manufacturing process history information of the raw material S1 to the location information mark M. In other words, in the linking step ST43 of the analysis method according to the fourth embodiment, unlike the linking step ST23 of the analysis method according to the second embodiment, in addition to the location information of defects present in the raw material S1 detected in the raw material inspection step ST41, the type of defects present in the raw material S1, and the execution location of the raw material inspection step ST41 in which the defects present in the raw material S1 were detected, the manufacturing process history information of the raw material S1 is linked to and stored in the location information mark M. Examples of manufacturing process history information include the values ​​of various parameters set or measured in the manufacturing process of the raw material S1, such as the stretching step, adhesive step, and bonding step.

[0080] [Analysis process ST46] In the analysis step ST46, similar to the analysis step ST26 of the analysis method according to the second embodiment, the position information acquired in the reading step ST45 is input to the control analysis device 4, and the control analysis device 4 identifies the position information of defects present in the sheet product S2 in the sheet product inspection step ST44, based on the position information of the sheet product S2 in the raw material S1 acquired in the reading step ST45. Then, in the analysis step ST46, similar to the analysis step ST26 of the analysis method according to the second embodiment, the control analysis device 4 is used to analyze the tendencies related to the occurrence of defects based on the positional information of the defects present in the identified sheet-like product S2 on the raw material S1. However, analysis step ST46 differs from analysis step ST26 in the following respects.

[0081] In the combination identification procedure ST461 of the analysis step ST46, it is determined whether the location information of the defects present in the identified sheet-like product S2 on the raw material S1 matches the location information of the defects present in the raw material S1 that is stored in association with the location information mark M read in the reading step ST45. This is the same as the combination identification procedure ST261 of the analysis step ST26 of the analysis method according to the second embodiment. However, the combination identification procedure ST461 of the fourth embodiment differs from the combination identification procedure ST261 of the second embodiment in that, when a match is found, the control analysis device 4 identifies a combination of the type of defect present in the raw material S1, which is stored in association with the position information mark M read in the reading process ST45, the execution location of the raw material inspection process ST41 in which the defect present in the raw material S1 was detected, and the manufacturing process history information of the raw material S1. In other words, unlike the combination identification procedure ST261, the combination identification procedure ST461 also identifies the manufacturing process history information of the raw material S1.

[0082] In the extraction condition determination procedure ST462 of the analysis process ST46, similar to the extraction condition determination procedure ST262 of the analysis process ST26 of the analysis method according to the second embodiment, the control analysis device 4 determines the extraction conditions based on the combination of type and execution location identified in the combination identification procedure ST461. In other words, the manufacturing process history information of the identified raw material S1 is not used when determining the extraction conditions. Then, in the indicator output procedure ST463 of the analysis process ST46, the control analysis device 4 outputs the manufacturing process history information identified in the combination identification procedure ST461 of raw material S1 containing defects that conform to the extraction conditions determined in the extraction condition determination procedure ST462, from among all the defects present in the raw material S1 detected in the raw material inspection process ST41, as an indicator representing the trend related to the occurrence of defects. For example, in the case of Table 1 mentioned above, the combination of "foreign matter" and "stretching process" is determined as the extraction condition, and the manufacturing process history information for raw material S1 containing defects that meet this extraction condition is output as an indicator representing the trend related to defect occurrence. Similarly, in the case of Table 2 mentioned above, the combination of "foreign matter" and "stretching process" and the combination of "dent" and "stretching process" are determined as the extraction conditions, and the manufacturing process history information for raw material S1 containing defects that meet this extraction condition is output as an indicator representing the trend related to defect occurrence.

[0083] As described above, according to the analysis method of the fourth embodiment, the identified manufacturing process history information of the raw material S1 containing defects that meet the extraction criteria is output as an index representing the trend related to the occurrence of defects. Therefore, it is expected that the trend in the manufacturing process history information of the raw material S1 in which defects may be detected in the sheet-like product S2 can be appropriately grasped.

[0084] <Fifth Embodiment> Figure 10 is a flowchart showing the schematic steps of the analysis method according to the fifth embodiment. As shown in Figure 10, the analysis method according to the fifth embodiment includes a raw material inspection step ST51, a position information marking step ST52, a linking step ST53, a sheet-like product inspection step ST54, a reading step ST55, and an analysis step ST56. The analysis method according to the fifth embodiment, like the analysis method according to the second embodiment, is a method that utilizes the inspection results of the raw material S1. The contents of the raw material inspection step ST51, position information marking step ST52, sheet-like product inspection step ST54, and reading step ST55 of the analysis method according to the fifth embodiment are the same as the contents of the raw material inspection step ST21, position information marking step ST22, sheet-like product inspection step ST24, and reading step ST25 of the analysis method according to the second embodiment. For this reason, the following will mainly describe the differences between the linking step ST53 and the analysis step ST56 from the linking step ST23 and the analysis step ST26 of the analysis method according to the second embodiment, and will omit detailed descriptions of other steps.

[0085] [Linking process ST53] In the linking process ST53, the control analysis device 4 links and stores the location information of defects present in the raw material S1 before cutting, which were detected in the raw material inspection process ST51, the type of defects present in the raw material S1, the execution location of the raw material inspection process ST51 in which the defects present in the raw material S1 were detected (the location of the inspection device 1), and the manufacturing process history information of the raw material S1 to the location information mark M. In other words, in the linking step ST53 of the analysis method according to the fifth embodiment, unlike the linking step ST23 of the analysis method according to the second embodiment, in addition to the location information of defects present in the raw material S1 detected in the raw material inspection step ST51, the type of defects present in the raw material S1, and the execution location of the raw material inspection step ST51 in which the defects present in the raw material S1 were detected, the manufacturing process history information of the raw material S1 is linked to and stored in the location information mark M. This is the same as the linking step ST43 of the analysis method according to the fourth embodiment.

[0086] [Analysis process ST56] In the analysis step ST56, similar to the analysis step ST26 of the analysis method according to the second embodiment, the position information acquired in the reading step ST55 is input to the control analysis device 4, and the control analysis device 4 identifies the position information of defects present in the sheet product S2 in the sheet product inspection step ST54 based on the position information of the sheet product S2 in the raw material S1 acquired in the reading step ST55. Then, in the analysis step ST56, similar to the analysis step ST26 of the analysis method according to the second embodiment, the control analysis device 4 is used to analyze the tendencies related to the occurrence of defects based on the positional information of the defects present in the identified sheet-like product S2 on the raw material S1. However, analysis step ST56 differs from analysis step ST26 in the following respects.

[0087] In the combination identification procedure ST561 of the analysis step ST56, it is determined whether the location information of the defects present in the identified sheet-like product S2 on the raw material S1 matches the location information of the defects present in the raw material S1 that is stored in association with the location information mark M read in the reading step ST55. This is the same as the combination identification procedure ST261 of the analysis step ST26 of the analysis method according to the second embodiment. However, the combination identification procedure ST561 of the fifth embodiment differs from the combination identification procedure ST261 of the second embodiment in that, when a match is found, the control analysis device 4 identifies a combination of the type of defect present in the raw material S1, which is stored in association with the position information mark M read in the reading process ST55, the execution location of the raw material inspection process ST51 in which the defect present in the raw material S1 was detected, and the manufacturing process history information of the raw material S1. In other words, unlike the combination identification procedure ST261, the combination identification procedure ST561 also identifies the manufacturing process history information of the raw material S1. This point is the same as the combination identification procedure ST461 of the analysis method according to the fourth embodiment.

[0088] In the extraction condition determination procedure ST562 of the analysis process ST56, the control analysis device 4 determines the extraction conditions based on the combination of type and execution location identified in the combination identification procedure ST561, similar to the extraction condition determination procedure ST262 of the analysis method according to the second embodiment. In other words, the manufacturing process history information of the identified raw material S1 is not used when determining the extraction conditions. Furthermore, in the analysis step ST56, unlike the analysis step ST26 of the analysis method according to the second embodiment, the control analysis device 4 has a determination procedure ST563 in which it makes a determination using a learning model as a determination of the tendency related to the occurrence of defects. The learning model is stored in the control analysis device 4 in an updatable manner. The learning model may be generated by supervised learning using known input and output combinations as training data, or it may be generated by unsupervised learning. As the learning model, any known learning model such as a neural network or a support vector machine may be used as appropriate.

[0089] Specifically, in the judgment procedure ST563, among all defects present in the raw material S1 detected in the raw material inspection process ST51, the manufacturing process history information identified in the combination identification procedure ST561 of raw material S1 containing defects that meet the extraction conditions determined in the extraction condition determination procedure ST562 is input to the learning model, and the learning model outputs a judgment result (probability of detection, etc.) of whether or not the defects present in the raw material S1 that meet the extraction conditions are detected as defects in the sheet-like product S2.

[0090] As described above, according to the analysis method of the fifth embodiment, in the determination procedure ST563, the identified manufacturing process history information of the raw material S1 containing defects that meet the extraction conditions is input to the learning model stored in the control analysis device 4, and the learning model outputs a determination result of whether or not the defects in the raw material S1 that meet the extraction conditions are detected as defects in the sheet-like product S2. Therefore, it is possible to evaluate (determine) the possibility of defects being detected (occurring) in the sheet-like product S2 according to the manufacturing process history information of the raw material S1.

[0091] As described above, the analysis method according to this embodiment (first to fifth embodiment) makes it possible to thoroughly investigate the cause of defects in the sheet-like product S2 and contribute significantly to improving the yield of the sheet-like product S2.

[0092] In this embodiment, the example described is the production of multiple sheet-like products S2 by unwinding, transporting, and cutting a long roll of raw material S1 wound in a roll shape. However, the application of the present invention is not limited to this. It can also be applied to the production of multiple sheet-like products S2 by unwinding and transporting a long roll of raw material S1 wound in a roll shape, marking it with position information marks M, and then cutting a large sheet-like intermediate (a sheet-like intermediate larger in dimensions than the multiple sheet-like products S2), and then cutting this intermediate. The intermediate can be cut while being transported by a conveyor R3 or the like, or it can be cut while placed on a predetermined cutting table.

[0093] Furthermore, although this embodiment describes the case where a position information mark M is marked on the raw material S1 before cutting, the present invention is not limited to this, and it is also possible to mark the position information mark M on intermediate products cut from the raw material S1 or on multiple sheet-like products S2 after cutting.

[0094] Figure 11 shows a schematic configuration of a manufacturing apparatus for performing an analysis method according to a modified version of the present invention (a configuration in which positional information marks M are marked on multiple sheet-like products S2 after cutting). In Figure 11, the same reference numerals are used for components similar to those in the manufacturing apparatus 100a shown in Figure 1. As shown in Figure 11, in the modified manufacturing apparatus 100a', the marking device 2 is positioned on the conveyor R3 downstream of the cutting device 3 (downstream in the direction of transport of the sheet-like product S2), and before the sheet-like product S2 is collected, the marking device 2 marks the sheet-like product S2 with a position information mark M.

[0095] Specifically, for example, the control and analysis device 4 can calculate the timing at which a predetermined portion of the raw material S1 reaches the marking device 2 after being cut into the sheet-like product S2, based on the separation distance L2' in the X direction between the feed roller R1 and the marking device 2, the amount of raw material S1 being transported measured by an encoder (not shown) attached to the nip roller R2, and the amount of sheet-like product S2 being transported measured by an encoder (not shown) attached to a roller on the conveyor R3. Then, for example, the control and analysis device 4 controls the marking device 2 so that transparent ink is sprayed from the nozzle of the marking device 2 corresponding to the Y coordinate of the predetermined XY coordinate when a portion having a predetermined XY coordinate (in the example shown in Figure 3, the portion where the position information mark M within each rectangle is marked) with respect to the planned cutting line CL of the raw material S1 (the cutting line that becomes the edge of the sheet-like product S2 after being cut into the sheet-like product S2) reaches the marking device 2. As a result, in the example shown in Figure 3, a position information mark M will be marked within each rectangle (within the sheet-like product S2).

[0096] Furthermore, as a method for marking the location information mark M, it is also possible to use the identification information described in Japanese Patent Publication No. 2005-114624. Specifically, as described in the above-mentioned publication, identification information representing the position of the raw material S1 in the transport direction is recorded at the widthwise (Y-direction) end of the raw material S1 (the end not used as the sheet-like product S2). For example, this identification information is read by a predetermined reading device positioned immediately before the cutting device 3 and input to the control and analysis device 4. Based on the read identification information and the amount of sheet-like product S2 transported measured by an encoder (not shown) attached to a roller on the conveyor R3, the control and analysis device 4 recognizes the position of the sheet-like product S2 in the transport direction on the raw material S1 after cutting. It can then calculate the timing at which a portion having predetermined XY coordinates (in the example shown in Figure 3, the portion where position information marks M within each rectangle are marked) of the sheet-like product S2 being transported sequentially after cutting reaches the marking device 2. The control and analysis device 4 can also control the marking device 2 to spray transparent ink from the nozzle of the marking device 2 corresponding to the Y coordinate of the predetermined XY coordinates at the above timing.

[0097] Furthermore, although this embodiment has described the case in which the position information marking process ST1, the cutting process ST2, and the recovery process ST3 are performed in a single manufacturing apparatus 100a (in other words, a single manufacturing line), the present invention is not limited to this. For example, after performing the position information marking process ST1, it is also possible to adopt a configuration in which the raw material S1 marked with the position information mark M is first wound onto a winding roller (not shown), transported to another manufacturing line, and then the cutting process ST2 and the recovery process ST3 are performed. In other words, it is also possible to divide the series of processes of the position information marking process ST1, the cutting process ST2, and the recovery process ST3 between the position information marking process ST1 and the cutting process ST2.

[0098] Furthermore, while the second to fifth embodiments were described using as an example the case in which the raw material inspection process (ST21, ST31, ST41, ST51) and the position information marking process (ST22, ST32, ST42, ST52) are executed in that order, the present invention is not limited to this, and it is also possible to adopt a configuration in which the position information marking process is executed first, followed by the raw material inspection process.

[0099] Furthermore, although this embodiment has described the case in which the analysis steps (ST14, ST26, ST36, ST46, ST56) are performed independently and separately, the present invention is not limited to this. It is also possible to configure the control analysis device 4 to be capable of performing at least two or more of the analysis steps described in each embodiment, and to adopt a configuration that allows the user to select which analysis step to perform. [Explanation of symbols]

[0100] 1. Inspection device 2. Marking device 3...Cutting device 4. Control and analysis device 100 System 100a...Manufacturing equipment F... Disadvantages M...Location information mark S1...original fabric S2... Sheet-type product

Claims

1. A method for analyzing the tendency of defects to occur in multiple sheet-like products, which are manufactured by unwinding, transporting, and cutting a long roll of raw material, or by unwinding, transporting, cutting a large sheet-like intermediate, and then cutting the intermediate, the method being used to analyze the tendencies related to the occurrence of defects, based on defects that occur in these products. A position information marking step, in which position information marks, which are marks representing positional information in the longitudinal and width directions on the raw material roll, are marked on the raw material roll, the intermediate product, or a plurality of sheet-like products before cutting or before cutting out the intermediate product, such that position information marks are present on each of the plurality of sheet-like products, A sheet-like product inspection process for inspecting multiple sheet-like products, A reading step in which the position information of the sheet product in the raw material is obtained by reading the position information mark marked on the sheet product in which a defect was detected in the sheet product inspection step, Based on the positional information of the sheet-like product on the raw material obtained in the reading step, the positional information of defects present in the sheet-like product detected in the sheet-like product inspection step is identified on the raw material, and based on the identified positional information of defects present in the sheet-like product on the raw material, an analysis step is performed to analyze the trends related to the occurrence of defects. A raw material inspection step for inspecting the raw material before cutting or before cutting out the intermediate, The system includes a linking step that stores, in association with a location information mark, the location information of defects present in the raw material before cutting or before cutting the intermediate, the type of defect present in the raw material, and the location where the raw material inspection step in which the defect was detected was performed. The aforementioned analysis step is A combination identification procedure that determines whether the location information of the defect present in the sheet-like product identified in the sheet-like product matches the location information of the defect present in the sheet-like product stored in association with the location information mark read in the reading step, and if they match, identifies a combination of the type of defect present in the sheet-like product stored in association with the location information mark read in the reading step and the location where the sheet-like product inspection step in which the defect present in the sheet-like product was detected was performed. A procedure for determining extraction conditions, which determines extraction conditions based on the combination of type and execution location identified in the above-mentioned combination identification procedure, A map image generation procedure generates a map image in which the location of defects in the raw material is plotted, based on the location information of defects in the raw material that meet the extraction conditions determined in the extraction condition determination procedure, among all defects present in the raw material detected in the raw material inspection process. A matching image generation procedure that generates a matching image by pattern matching the aforementioned map image with a pre-prepared template image, wherein the matching image is an image in which only the locations of the defects plotted on the map image that match the locations of the defects in the template image are plotted, and this matching image is used to represent the trend related to the occurrence of defects. A method for analyzing the tendency of defects to occur, which has the following characteristics.

2. In the combination identification procedure, for each of the multiple sheet products in which defects were detected in the sheet product inspection process, a plurality of combinations of the type and the execution location are identified. In the extraction condition determination procedure, a majority of the combinations of the multiple types and execution locations identified in the combination identification procedure are determined to be the extraction conditions. A method for analyzing the occurrence trend of defects as described in claim 1.

3. In the combination identification procedure, for each of the multiple sheet products in which defects were detected in the sheet product inspection process, a plurality of combinations of the type and the execution location are identified. In the extraction condition determination procedure, all of the multiple combinations of the type and execution location identified in the combination identification procedure are determined as the extraction conditions. A method for analyzing the occurrence trend of defects as described in claim 1.

4. In the combination identification procedure, for each of the multiple sheet products in which defects were detected in the sheet product inspection process, a plurality of combinations of the type and the execution location are identified. The analysis step includes an index output procedure that, if the number of combinations of the plurality of types and execution locations identified in the combination identification procedure is less than half of the total number of defects detected in the sheet product inspection step for the plurality of sheet products, outputs that the number is less than half as an index representing a trend related to the occurrence of defects, without executing the extraction condition determination procedure, the map image generation procedure, and the matching image generation procedure. A method for analyzing the occurrence trend of defects as described in claim 1.

5. A method for analyzing the tendency of defects to occur, based on defects that occur in a plurality of sheet-like products manufactured by unwinding, transporting, and cutting a long roll of raw material wound in a roll, or by unwinding, transporting, cutting out a large sheet-like intermediate, and then cutting the intermediate, the method being used to analyze the tendency related to the occurrence of defects. A position information marking step, in which position information marks, which are marks representing positional information in the longitudinal and width directions on the raw material roll, are marked on the raw material roll, the intermediate product, or a plurality of sheet-like products before cutting or before cutting out the intermediate product, such that position information marks are present on each of the plurality of sheet-like products, A sheet-like product inspection process for inspecting multiple sheet-like products, A reading step in which the position information of the sheet product in the raw material is obtained by reading the position information mark marked on the sheet product in which a defect was detected in the sheet product inspection step, Based on the positional information of the sheet-like product on the raw material obtained in the reading step, the positional information of defects present in the sheet-like product detected in the sheet-like product inspection step is identified on the raw material, and based on the identified positional information of defects present in the sheet-like product on the raw material, an analysis step is performed to analyze the trends related to the occurrence of defects. A raw material inspection step for inspecting the raw material before cutting or before cutting out the intermediate, The system includes a linking step that stores, in association with the location information marks, the location information of defects present in the raw material before cutting or before cutting the intermediate, the type of defects present in the raw material, the location where the raw material inspection step in which the defects were detected was performed, and the manufacturing process history information of the raw material. The aforementioned analysis step is A combination identification procedure that determines whether the location information of the defect present in the sheet-like product identified in the sheet-like product matches the location information of the defect present in the sheet-like product stored in association with the location information mark read in the reading step, and if they match, identifies a combination of the type of defect present in the sheet-like product stored in association with the location information mark read in the reading step, the location where the sheet-like product inspection step in which the defect was detected was performed, and the manufacturing process history information of the sheet-like product. A procedure for determining extraction conditions, which determines extraction conditions based on the combination of type and execution location identified in the above-mentioned combination identification procedure, An indicator output procedure for outputting, as an indicator representing the trend related to defect occurrence, the manufacturing process history information identified in the combination identification procedure for the raw materials that have defects that match the extraction conditions determined in the extraction condition determination procedure, out of all defects present in the raw materials detected in the raw material inspection procedure, and which are found to be defects that match the extraction conditions determined in the extraction condition determination procedure, A method for analyzing the tendency of defects to occur, which has the following characteristics.

6. In the combination identification procedure, for each of the multiple sheet products in which defects were detected in the sheet product inspection process, a plurality of combinations of the type and the execution location are identified. In the extraction condition determination procedure, a majority of the combinations of the multiple types and execution locations identified in the combination identification procedure are determined to be the extraction conditions. A method for analyzing the tendency of defects to occur as described in claim 5.

7. A method for analyzing the tendency of defects to occur in a plurality of sheet-like products, which are manufactured by unwinding, transporting, and cutting a long roll of raw material wound in a roll, or by unwinding, transporting, cutting out a large sheet-like intermediate, and then cutting the intermediate, based on defects that occur in the sheet-like products, the method being used to analyze the tendency related to the occurrence of defects, A position information marking step, in which position information marks, which are marks representing positional information in the longitudinal and width directions on the raw material roll, are marked on the raw material roll, the intermediate product, or a plurality of sheet-like products before cutting or before cutting out the intermediate product, such that position information marks are present on each of the plurality of sheet-like products, A sheet-like product inspection process for inspecting multiple sheet-like products, A reading step in which the position information of the sheet product in the raw material is obtained by reading the position information mark marked on the sheet product in which a defect was detected in the sheet product inspection step, Based on the positional information of the sheet-like product on the raw material obtained in the reading step, the positional information of defects present in the sheet-like product detected in the sheet-like product inspection step is identified on the raw material, and based on the identified positional information of defects present in the sheet-like product on the raw material, an analysis step is performed to analyze the trends related to the occurrence of defects. A raw material inspection step for inspecting the raw material before cutting or before cutting out the intermediate, The system includes a linking step that stores, in association with the location information marks, the location information of defects present in the raw material before cutting or before cutting the intermediate, the type of defects present in the raw material, the location where the raw material inspection step in which the defects were detected was performed, and the manufacturing process history information of the raw material. The aforementioned analysis step is A combination identification procedure that determines whether the location information of the defect present in the sheet-like product identified in the sheet-like product matches the location information of the defect present in the sheet-like product stored in association with the location information mark read in the reading step, and if they match, identifies a combination of the type of defect present in the sheet-like product stored in association with the location information mark read in the reading step, the location where the sheet-like product inspection step in which the defect was detected was performed, and the manufacturing process history information of the sheet-like product. A procedure for determining extraction conditions, which determines extraction conditions based on the combination of type and execution location identified in the above-mentioned combination identification procedure, The system includes a determination procedure that uses a learning model to determine the tendencies related to the occurrence of defects, In the determination procedure, the manufacturing process history information identified in the combination identification procedure of the raw materials containing defects that meet the extraction conditions determined in the extraction condition determination procedure, from among all defects present in the raw materials detected in the raw material inspection step, is input to the learning model, and the learning model outputs a determination result of whether or not the defects present in the raw materials that meet the extraction conditions are detected as defects in the sheet-like product. Method for analyzing the tendency of defects to occur.

Images