Manufacturing line abnormality determination device, manufacturing line abnormality determination system, manufacturing line abnormality determination method and program

Abstract

To easily identify and confirm a product that may be defective.SOLUTION: An anomaly determination device of a manufacturing line is configured to: acquire a manufacturing line moving image that is a moving image including at least part of a manufacturing line for manufacturing a product; acquire an identification code image that is an image including an identification code given to the product; determine the presence / absence of anomaly of the manufacturing line based on the acquired manufacturing line moving image; acquire an imaging time of the manufacturing line moving image; and record an imaging time of the identification code image. When it is determined that anomaly in the manufacturing line occurs, in order to identify the product, which may be defective, included in the manufacturing line moving image, the acquired identification code image, the acquired imaging time of the manufacturing line moving image, and the acquired imaging time of the identification code image are used.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a manufacturing line abnormality determination device, a manufacturing line abnormality determination system, a manufacturing line abnormality determination method, and a program. [Background technology]

[0002] Conventionally, products have been manufactured using a manufacturing line. For example, in the technology described in Patent Document 1, a bag-in-box product is manufactured on a manufacturing line by filling an inner bag with a spout, attaching a cap to the spout, and storing the inner bag in an outer box. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-052582 Summary of the Invention [Problem to be solved by the invention]

[0004] The present inventors have conducted extensive research to examine the causes and trends of defective products manufactured on production lines such as the production line described in Patent Document 1. As a result, they have found that the likelihood of defective products occurring increases when the movement of the production units that manufacture products on the production line differs from the predetermined normal movement. They have also found that the likelihood of defective products occurring increases when the movement of products manufactured on the production line differs from the predetermined normal movement. They have also found that the likelihood of defective products occurring increases when the production line operators make unusual movements (for example, movements to check the production line, etc.). Conventionally, when a defective product is found, the production line monitoring video at the time when the product produced on the production line became defective is checked, and this is used to investigate the cause of the defective product.

[0005] Through extensive research, the inventors have discovered that a computer can determine, without a human being having to make a judgment, whether the movement of a manufacturing unit that manufactures products on a production line differs from a predetermined normal movement, based on video images for monitoring the production line. The inventors have also discovered that a computer can determine, without a human being having to make a judgment, whether the movement of a product manufactured on a production line differs from a predetermined normal movement, based on video images for monitoring the production line. The inventors have also discovered that a computer can determine, without a human being having to make a judgment, whether an operator on the production line is behaving abnormally, based on video images for monitoring the production line. Furthermore, through extensive research, the inventors have discovered that by recording the time when a product manufactured on a production line becomes defective (more specifically, the time when the video image for monitoring the production line is captured), as well as the time when an identification code image of an identification code attached to a product on the production line is captured, it becomes much easier to identify and confirm products that may have become defective.

[0006] In other words, the present invention aims to provide a manufacturing line abnormality determination device, a manufacturing line abnormality determination system, a manufacturing line abnormality determination method, and a program that can easily identify and confirm products that may have become defective. [Means for solving the problem]

[0007] One aspect of the present invention includes a video image acquisition unit that acquires a production line video image, which is a video image including at least a portion of a production line that manufactures products; an identification code image acquisition unit that acquires an identification code image, which is an image including an identification code added to the product; an abnormality determination unit that determines whether or not there is an abnormality in the production line based on the production line video image acquired by the video image acquisition unit; a video image capture time acquisition unit that acquires the image capture time of the production line video image; and an identification code image capture time acquisition unit that acquires the image capture time of the identification code image.When the abnormality determination unit determines that an abnormality has occurred in the production line, the identification code image acquired by the identification code image acquisition unit, the image capture time of the production line video image acquired by the video image capture time acquisition unit, and the image capture time of the identification code image acquired by the identification code image capture time acquisition unit are used to identify the product included in the production line video image that may have become a defective product. and an OCR determination unit that identifies the identification code included in the identification code image by performing OCR (Optical Character Recognition) determination on the identification code image acquired by the identification code image acquisition unit, and if the OCR determination unit can identify the identification code included in the identification code image, the identification code identified by the OCR determination unit is used to identify the product that may have become a defective product, and if the OCR determination unit cannot identify the identification code included in the identification code image, the identification code image is used to identify the product that may have become a defective product. This is an abnormality detection device for a manufacturing line.

[0008] One aspect of the present invention is an abnormality determination system for a production line, including the abnormality determination device, a first imaging device that captures the production line moving image, and a second imaging device that captures the identification code image.

[0009] One aspect of the present invention includes a video image acquisition step of acquiring a production line video image, which is a video image including at least a part of a production line that manufactures products; an identification code image acquisition step of acquiring an identification code image, which is an image including an identification code added to the product; an abnormality determination step of determining whether or not there is an abnormality in the production line based on the production line video image acquired in the video image acquisition step; a video image capture time acquisition step of acquiring the image capture time of the production line video image; and an identification code image capture time acquisition step of acquiring the image capture time of the identification code image. When it is determined in the abnormality determination step that an abnormality has occurred in the production line, the identification code image acquired in the identification code image acquisition step, the image capture time of the production line video image acquired in the video image capture time acquisition step, and the image capture time of the identification code image acquired in the identification code image capture time acquisition step are used to identify the product included in the production line video image that may have become a defective product. and further comprising an OCR step of identifying the identification code included in the identification code image by performing OCR (Optical Character Recognition) judgment on the identification code image acquired by the identification code image acquisition step, and if the OCR step can identify the identification code included in the identification code image, the identification code identified by the OCR step is used to identify the product that may have become a defective product, and if the OCR step cannot identify the identification code included in the identification code image, the identification code image is used to identify the product that may have become a defective product. This is a method for determining abnormalities in a production line.

[0010] One aspect of the present invention is a program for causing a computer to execute a video image acquisition step of acquiring a production line video image, which is a video image including at least a part of a production line where products are manufactured; an identification code image acquisition step of acquiring an identification code image, which is an image including an identification code added to the product; an abnormality determination step of determining whether or not there is an abnormality in the production line based on the production line video image acquired in the video image acquisition step; a video image capture time acquisition step of acquiring the image capture time of the production line video image; and an identification code image capture time acquisition step of acquiring the image capture time of the identification code image, wherein when it is determined in the abnormality determination step that an abnormality has occurred in the production line, the identification code image acquired in the identification code image acquisition step, the image capture time of the production line video image acquired in the video image capture time acquisition step, and the image capture time of the identification code image acquired in the identification code image capture time acquisition step are used to identify the product included in the production line video image that may have become a defective product. and causing a computer to execute an OCR (Optical Character Recognition) step of identifying the identification code included in the identification code image by performing OCR judgment on the identification code image acquired in the identification code image acquisition step, and if the OCR step can identify the identification code included in the identification code image, the identification code identified in the OCR step is used to identify the product that may have become a defective product, and if the OCR step cannot identify the identification code included in the identification code image, the identification code image is used to identify the product that may have become a defective product. It is a program. [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a manufacturing line abnormality determination device, a manufacturing line abnormality determination system, a manufacturing line abnormality determination method, and a program that can easily identify and confirm products that may have become defective. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing an example of a manufacturing line abnormality determination system S to which a manufacturing line abnormality determination device 1 according to a first embodiment is applied. [Figure 2] 2 is a diagram showing an example of a production line PL in which the presence or absence of an abnormality is determined by the abnormality determination system S shown in FIG. 1. FIG. [Figure 3] FIG. 3 is a diagram showing a specific example of the production line PL shown in FIG. 2. [Figure 4] 4A and 4B are diagrams showing examples of images captured by imaging devices S1 and S2 in the example shown in FIG. [Figure 5] 5 is a diagram showing an example of a data flow in the abnormality determination system S for the production line shown in FIGS. 1 to 4. FIG. [Figure 6] 4 is a flowchart illustrating an example of processing executed in the abnormality determination device 1 of the manufacturing line according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] First Embodiment DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, with reference to the accompanying drawings, an embodiment of a manufacturing line abnormality determination device, a manufacturing line abnormality determination system, a manufacturing line abnormality determination method, and a program according to the present invention will be described.

[0014] Fig. 1 is a diagram showing an example of a production line abnormality determination system S to which the production line abnormality determination device 1 of the first embodiment is applied. Fig. 2 is a diagram showing an example of a production line PL in which the presence or absence of an abnormality is determined by the abnormality determination system S shown in Fig. 1. 1 and 2, a product P is manufactured in a manufacturing line PL. An identification code PD is attached to the product P. The manufacturing line PL includes a transport unit PL1 (e.g., a conveyor) that transports the product P, and a manufacturing unit PL2 that manufactures the product P (i.e., performs processing to complete the product P). The abnormality determination system S includes an imaging device S1, an imaging device S2, and an abnormality determination device 1. The imaging device S1 captures images of a production line moving image LM (see FIG. 4(B)), which is a moving image including at least a portion of a production line PL that manufactures a product P. The imaging device S1 includes an imaging unit S1A and a recording unit S1B. The imaging unit S1A captures the production line moving image LM. Specifically, the imaging unit S1A generates data of the production line moving image LM (production line moving image data). The recording unit S1B records the image capture time of the production line moving image LM captured by the imaging unit S1A. Specifically, the recording unit S1B generates data indicating the image capture time of the production line moving image LM (production line moving image capture time data). The imaging device S2 captures an identification code image DM (see FIG. 4A), which is an image including the identification code PD added to the product P. The imaging device S2 includes an imaging unit S2A and a recording unit S2B. The imaging unit S2A captures the identification code image DM. Specifically, the imaging unit S2A generates data of the identification code image DM (identification code image data). The recording unit S2B records the image capture time of the identification code image DM captured by the imaging unit S2A. Specifically, the recording unit S2B generates data indicating the image capture time of the identification code image DM (identification code image capture time data).

[0015] The abnormality judgment device 1 includes a moving image acquisition unit 1A, an identification code image acquisition unit 1B, an abnormality judgment unit 1C, a moving image capture time acquisition unit 1D, an identification code image capture time acquisition unit 1E, an OCR (Optical Character Recognition) judgment unit 1F, and an alarm issuance unit 1G. The moving image acquisition unit 1A acquires the production line moving image LM captured by the imaging unit S1A of the imaging device S1. Specifically, the moving image acquisition unit 1A acquires the production line moving image data generated by the imaging unit S1A. The identification code image acquiring unit 1B acquires the identification code image DM captured by the imaging unit S2A of the imaging device S2. In detail, the identification code image acquiring unit 1B acquires the identification code image data generated by the imaging unit S2A.

[0016] The abnormality determination unit 1C determines whether or not an abnormality exists in the production line PL based on the production line moving image LM acquired by the moving image acquisition unit 1A. Specifically, the abnormality determination unit 1C learns the determination criteria used to determine whether or not an abnormality exists in the production line PL. Specifically, the abnormality determination unit 1C performs supervised learning using training data that is a combination of learning production line moving images, which are moving images including at least a portion of the production line PL, and information included in the learning production line moving images that indicates whether or not an abnormality has occurred in the production line PL. Therefore, when an abnormality occurs in the production line PL, the abnormality determination unit 1C that has performed supervised learning can determine that an abnormality has occurred in the production line PL based on the production line moving image LM. Also, when an abnormality does not occur in the production line PL, the abnormality determination unit 1C that has performed supervised learning can determine that an abnormality has not occurred in the production line PL based on the production line moving image LM.

[0017] The video image capturing time acquisition unit 1D acquires the image capturing time of the production line video image LM recorded by the recording unit S1B of the imaging device S1. Specifically, the video image capturing time acquisition unit 1D acquires the production line video image capturing time data generated by the recording unit S1B. The identification code image capturing time acquiring unit 1E acquires the capturing time of the identification code image DM recorded by the recording unit S2B of the imaging device S2. In detail, the identification code image capturing time acquiring unit 1E acquires the identification code image capturing time data generated by the recording unit S2B. Therefore, in the example shown in Figures 1 and 2, if the abnormality judgment unit 1C determines that an abnormality has occurred on the production line PL, an operator, manager, etc. of the production line PL can identify a product P included in the production line moving image LM that may have become a defective product by using the imaging time of the production line moving image LM acquired by the imaging time acquisition unit 1D, the imaging time of the identification code image DM acquired by the identification code image imaging time acquisition unit 1E, the time required for the product P to move between the position of the product P included in the production line moving image LM and the position of the product P included in the identification code image DM, and the identification code PD added to the product P included in the identification code image DM acquired by the identification code image acquisition unit 1B. In other words, in the example shown in Figures 1 and 2, if the abnormality determination unit 1C determines that an abnormality has occurred in the production line PL, for example, an operator or manager of the production line PL can easily identify and confirm products P that may have become defective.

[0018] In the example shown in FIGS. 1 and 2, the OCR determination unit 1F identifies the identification code PD included in the identification code image DM by performing OCR determination on the identification code image DM acquired by the identification code image acquisition unit 1B. If the OCR determination unit 1F can identify the identification code PD included in the identification code image DM, the identification code PD identified by the OCR determination unit 1F is used to identify products P that may have become defective when the abnormality determination unit 1C determines that an abnormality has occurred on the production line PL. On the other hand, if the OCR determination unit 1F cannot identify the identification code PD included in the identification code image DM, the identification code image DM is used to identify products P that may have become defective when the abnormality determination unit 1C determines that an abnormality has occurred on the production line PL. When the abnormality determination unit 1C determines that an abnormality has occurred in the production line PL, the alarm issuance unit 1G issues an alarm to, for example, an operator or manager of the production line PL. When the alarm issuance unit 1G issues an alarm, the production line PL may be stopped.

[0019] FIG. 3 is a diagram showing a specific example of the production line PL shown in FIG. In the example shown in FIG. 3, the product P manufactured in the production line PL is a bag-in-box B, and the production line PL is configured by a bag-in-box filling machine 11. The bag-in-box B has an outer box B1, an inner bag B2 stored in the outer box B1, a spout B21 attached to the inner bag B2, and a cap B3 attached to the spout B21. The bag-in-box filling machine 11 fills the bag-in-box B with contents. The manufacturing line PL (bag-in-box filling machine 11) includes a transport unit PL1 that transports the bag-in-box B and a manufacturing unit PL2 that manufactures the bag-in-box B (i.e., performs processing to complete the bag-in-box B). The manufacturing unit PL2 includes a filling section 112 that fills the inner bag B2 with contents, a capping mechanism that opens and closes the cap B3 on the spout B21 (see, for example, Patent Publication No. 2018-138464), and a cutting section 114 that cuts multiple inner bags B2 that are connected to each other into individual inner bags B2. The conveying unit PL1 includes an inner bag supply section 111 that supplies multiple inner bags B2 that are connected to each other to the filling section 112, and a lifting section 112D that raises and lowers a filling table 112C that supports the inner bag B2 when the contents are filled into the inner bag B2 by the filling section 112.

[0020] The filling section 112 includes a tank 112E, a pump 11B, a flow meter 11C, and a filling nozzle 112A. Tank 112E contains the contents to be filled into inner bag B2. Pump 11B supplies the contents contained in tank 112E to filling nozzle 112A. Flow meter 11C measures the flow rate of the contents supplied by pump 11B. The filling nozzle 112A is configured to be insertable into the spout B21, and fills the inner bag B2 with the contents supplied by the pump 11B.

[0021] The filling table 112C is lowered by the lifting section 112D as the amount of contents filled into the inner bag B2 increases. The capping mechanism opens the cap B3 before the filling unit 112 starts filling the contents, and closes the cap B3 after the filling unit 112 has finished filling the contents. The cutting unit 114 cuts the multiple inner bags B2, which are connected to each other via connecting parts having perforations, at the positions of the perforations to separate them into individual inner bags B2. The cutting unit 114 includes a cutter 114A and an elevating unit 114B that raises and lowers the cutter 114A. The inserting section 115 inserts the inner bag B2 that has been filled with the contents by the filling section 112 and cut by the cutting section 114 into the outer box B1.

[0022] Fig. 4 is a diagram showing an example of images captured by the imaging devices S1 and S2 in the example shown in Fig. 3. In detail, Fig. 4(A) shows an example of an identification code image DM (an image including the identification code PD added to the product P (more specifically, the inner bag B2)) captured by the imaging unit S2A of the imaging device S2 in the example shown in Fig. 3, and Fig. 4(B) shows an example of a production line moving image LM (a moving image including a part of the production line PL (more specifically, the bag-in-box filling machine 11)) captured by the imaging unit S1A of the imaging device S1 in the example shown in Fig. 3. In the example shown in FIGS. 3 and 4, the imaging section S2A of the imaging device S2 captures an identification code image DM (an image including the identification code PD affixed to the inner bag B2) from above (the upper side of FIG. 3) at a position upstream (the right side of FIG. 3) of the filling section 112 of the bag-in-box filling machine 11. The imaging section S1A of the imaging device S1 captures a production line video image LM, for example, from the front (the near side of FIG. 3) of the bag-in-box filling machine 11. The production line video image LM includes the product P (inner bag B2) being transported by the transport unit PL1 of the production line PL (e.g., the inner bag supply section 111, the lifting section 112D, etc.). The production line video image LM also includes the production unit PL2 (e.g., the filling section 112, the capping mechanism, the cutting section 114, etc.) that produces the product P (bag-in-box B).

[0023] FIG. 5 is a diagram showing an example of the flow of data in the abnormality determination system S for the production line shown in FIGS. In the example shown in FIG. 5, a control device (e.g., a programmable logic controller (PLC) of the filling machine main body) of the bag-in-box filling machine 11 outputs a detection trigger indicating the timing to capture an identification code image DM to an OCR determination unit 1F (e.g., an OCR determination controller control PLC) of the abnormality determination device 1. Next, based on the detection trigger, the OCR determination unit 1F outputs an image capture request for the identification code image DM to the imaging unit S2A of the imaging device S2. Next, the imaging unit S2A captures the identification code image DM based on the image capture request and outputs the identification code image data. Next, the OCR determination unit 1F identifies the identification code PD included in the identification code image DM by performing OCR determination on the identification code image DM (i.e., generates print data of the identification code PD). Furthermore, the OCR determination unit 1F stores the identification code image data in a storage device, which is used to identify a product P that may be defective if the identification code PD cannot be identified.

[0024] 5, the imaging unit S1A of the imaging device S1 captures a production line moving image LM and outputs the production line moving image data. Next, the abnormality determination unit 1C (e.g., a PC (Personal Computer) equipped with an AI (Artificial Intelligence) engine) determines whether or not there is an abnormality in the production line PL based on the production line moving image LM and outputs the determination result to the control device of the bag-in-box filling machine 11. The abnormality determination unit 1C also saves the production line moving image data. The bag-in-box filling machine 11 is equipped with an alarm lamp, a monitoring screen, etc. The control device, OCR determination unit 1F, and abnormality determination unit 1C of the bag-in-box filling machine 11 are connected to a host system PLC. The host system PLC is also connected via a LAN (Local Area Network) to a plurality of PCs in an office (i.e., located away from the bag-in-box filling machine 11) for viewing (monitoring) collected data.

[0025] In other words, in the example shown in Figure 5, AI detects in advance any unusual behavior of the machine (manufacturing unit PL2), product P, or person (operator of the manufacturing line PL) that tends to occur when product P (bag-in-box B) becomes defective, and this is used to investigate the cause of the defective product and prevent it from being released. In addition, by acquiring the individual number (identification code PD attached to product P) of each inner bag B2 as digital data using an OCR camera (imaging section S2A of imaging device S2) and linking it to video data (data from the manufacturing line video image LM), it is possible to quickly check the manufacturing history, leading to early discovery of the cause and identification of the extent of the impact. Furthermore, if a machine (manufacturing unit PL2) behaves differently than normal, the results of the abnormality determination unit 1C can be used to investigate the cause of the machine's (manufacturing unit PL2) failure and for preventive maintenance. If a product P (such as the inner bag B2 of a bag-in-box B) behaves differently than normal, an abnormality in the bag production process can be detected based on the results of the abnormality determination unit 1C. If a person (an operator of the manufacturing line PL) behaves differently than normal, this can be used as a tool to educate the person on incorrect behavior.

[0026] FIG. 6 is a flowchart for explaining an example of processing executed in the abnormality determination device 1 of the manufacturing line according to the first embodiment. In the example shown in FIG. 6, the identification code image acquisition unit 1B of the abnormality determination device 1 acquires an identification code image DM that is an image including the identification code PD added to the product P in step S11. In step S12, the identification code image capturing time acquisition unit 1E of the abnormality determination device 1 acquires the capturing time of the identification code image DM. In step S13, the moving image acquisition unit 1A of the abnormality determination device 1 acquires a production line moving image LM that is a moving image including at least a part of the production line PL on which the product P is manufactured. In step S14, the moving image capturing time acquisition unit 1D of the abnormality determination device 1 acquires the capturing time of the production line moving image LM. In step S15, the abnormality determination unit 1C of the abnormality determination device 1 determines whether or not there is an abnormality in the production line PL (whether or not there is a risk that the product P has become a defective product) based on the production line moving image LM acquired in step S13. If it is determined in step S15 that an abnormality has occurred on the production line PL, the identification code image DM acquired in step S11, the imaging time of the identification code image DM acquired in step S12, and the imaging time of the production line moving image LM acquired in step S14 are used to identify the product P included in the production line moving image LM that may have become a defective product.

[0027] As described above, in the manufacturing line abnormality determination system S to which the manufacturing line abnormality determination device 1 of the first embodiment is applied, by introducing AI testing, the abnormality determination section 1C of the abnormality determination device 1 detects behavior that differs from the normal normal operation of the machine (manufacturing unit PL2), product P, or person (operator of the manufacturing line PL), and the alarm issuance section 1G of the abnormality determination device 1 issues an alarm, thereby preventing the outflow of defective products. Furthermore, in the abnormality determination system S for a production line of the first embodiment, an OCR camera (imaging section S2A of imaging device S2) images the entire film (the entire inner bag B2 to which the film lot number (identification code PD) is attached), and an identification code image acquisition section 1B of the abnormality determination device 1 acquires an image of the entire film (an identification code image DM that is an image including the identification code PD). An OCR determination section 1F of the abnormality determination device 1 identifies the identification code PD (film lot number) included in the identification code image DM by performing OCR determination on the identification code image DM. As a result, by linking the identification code PD (film lot number) attached to the product P included in the production line video image LM (i.e., the product P that may have become a defective product) with the identification code PD (film lot number) included in the identification code image DM captured by the OCR camera (the imaging section S2A of the imaging device S2), it is possible to easily identify the product P that may have become a defective product.

[0028] By connecting the management status via a network (LAN) as in the example shown in Figure 5, it becomes possible to monitor the equipment (production line PL) all at once, and the production line PL manager, etc. can check the production history and defective products without having to visit the production site (production line PL) of product P. It is difficult to isolate errors using only the camera for AI video inspection (imaging section S1A of imaging device S1), but by combining the camera for AI video inspection with an OCR camera (imaging section S2A of imaging device S2), it is possible to easily identify the cause of defects in each of the inner bags B2 of bag-in-box B.

[0029] As described above, the abnormality determination system S for a manufacturing line according to the first embodiment can improve the quality assurance level by suppressing the outflow of defective products P (bag-in-box B). In addition, when a defective product P (bag-in-box B) is found, the work of checking the manufacturing history can be shortened.

[0030] Second Embodiment A second embodiment of the manufacturing line abnormality determination device, manufacturing line abnormality determination system, manufacturing line abnormality determination method, and program of the present invention will be described below. The production line abnormality determination system S to which the production line abnormality determination device 1 of the second embodiment is applied is configured similarly to the production line abnormality determination system S of the first embodiment described above, except for the points described below. Therefore, the production line abnormality determination system S of the second embodiment can achieve the same effects as the production line abnormality determination system S of the first embodiment described above, except for the points described below.

[0031] A production line abnormality determination system S to which the production line abnormality determination device 1 of the second embodiment is applied is configured similarly to the production line abnormality determination system S shown in Fig. 1. The production line abnormality determination system S of the second embodiment determines whether or not there is an abnormality in the production line PL (see Fig. 2).

[0032] In the second embodiment of the abnormality determination device 1 for a manufacturing line, similar to the first embodiment of the abnormality determination device 1 for a manufacturing line, the abnormality determination unit 1C determines whether or not there is an abnormality in the manufacturing line PL based on the manufacturing line moving image LM acquired by the moving image acquisition unit 1A. In detail, as described above, in the manufacturing line abnormality determination device 1 of the first embodiment, the abnormality determination unit 1C learns the determination criteria used to determine whether or not an abnormality exists in the manufacturing line PL. On the other hand, in the manufacturing line abnormality determination device 1 of the second embodiment, the abnormality determination unit 1C does not learn the determination criteria used to determine whether or not there is an abnormality in the manufacturing line PL.

[0033] In the second embodiment of the manufacturing line abnormality determination device 1, when the manufacturing line moving image LM acquired by the moving image acquisition unit 1A includes an operator of the manufacturing line PL, the abnormality determination unit 1C determines that an abnormality has occurred (or is likely to occur) on the manufacturing line PL. This is because the operator of the production line PL is not normally within the imaging range of the imaging unit S1A of the imaging device S1, but approaches the production line PL when he or she senses an abnormality or malfunction in the production line PL.

[0034] In addition, in the second embodiment of the abnormality determination device 1 for a manufacturing line, if the movement of the product P included in the manufacturing line moving image LM acquired by the moving image acquisition unit 1A differs from the predetermined normal movement, the abnormality determination unit 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL. The product P normally moves at a constant speed or at a speed that changes periodically (a preset movement), but if an abnormality occurs in the production line PL, the movement of the product P will no longer be the preset movement.

[0035] Furthermore, in the manufacturing line abnormality determination device 1 of the second embodiment, when the movement of a manufacturing unit PL2 (e.g., the filling section 112, capping mechanism, cutting section 114, etc.) included in the manufacturing line moving image LM acquired by the moving image acquisition section 1A differs from the predetermined normal movement (e.g., when the manufacturing unit PL2 deteriorates and the operating speed of the manufacturing unit PL2 slows down), the abnormality determination section 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL. The manufacturing unit PL2 normally behaves in a preset manner, but in the event of an abnormality, such as when the manufacturing unit PL2 deteriorates, the manufacturing unit PL2 will no longer behave in the preset manner. In an example in which the second embodiment of the abnormality determination device 1 for a manufacturing line is applied to the bag-in-box filling machine 11 shown in Figure 3, for example, if the movement of any of the filling section 112, inner bag supply section 111, lifting section 112D, capping mechanism and cutting section 114 included in the manufacturing line moving image LM acquired by the moving image acquisition section 1A differs from the predetermined normal movement, the abnormality determination section 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL.

[0036] In the second embodiment of the manufacturing line abnormality judgment device 1, when the abnormality judgment unit 1C judges that an abnormality has occurred (or is likely to occur) in the manufacturing line PL, an operator, manager, etc. of the manufacturing line PL can identify the product P (bag-in-box B or inner bag B2) included in the manufacturing line moving image LM that may be defective by using the imaging time of the manufacturing line moving image LM acquired by the imaging time acquisition unit 1D, the imaging time of the identification code image DM acquired by the identification code image imaging time acquisition unit 1E, the time required for the product P to move between the position of the product P included in the manufacturing line moving image LM and the position of the product P included in the identification code image DM, and the identification code PD added to the product P included in the identification code image DM acquired by the identification code image acquisition unit 1B. In other words, in the abnormality judgment device 1 for a manufacturing line of the second embodiment, when the abnormality judgment unit 1C judges that an abnormality has occurred (or may have occurred) on the manufacturing line PL, an operator or manager of the manufacturing line PL, for example, can easily identify and confirm the product P (bag-in-box B or inner bag B2) that may have become a defective product.

[0037] <Third embodiment> A third embodiment of the production line abnormality determination device, production line abnormality determination system, production line abnormality determination method, and program of the present invention will be described below. The production line abnormality determination system S to which the production line abnormality determination device 1 of the third embodiment is applied is configured similarly to the production line abnormality determination system S of the first embodiment described above, except for the points described below. Therefore, the production line abnormality determination system S of the third embodiment can achieve the same effects as the production line abnormality determination system S of the first embodiment described above, except for the points described below.

[0038] A production line abnormality determination system S to which the production line abnormality determination device 1 of the third embodiment is applied is configured similarly to the production line abnormality determination system S shown in Fig. 1. The production line abnormality determination system S of the third embodiment determines whether or not there is an abnormality in the production line PL (see Fig. 2).

[0039] As described above, the manufacturing line abnormality determination system S to which the manufacturing line abnormality determination device 1 of the first embodiment is applied determines whether or not there is an abnormality in the manufacturing line PL (bag-in-box filling machine 11) of the bag-in-box B (product P). On the other hand, the production line abnormality determination system S to which the production line abnormality determination device 1 of the third embodiment is applied determines whether or not there is an abnormality in the production line PL of products P other than bag-in-box B (for example, automobile parts, processed foods, etc.).

[0040] In the production line abnormality determination system S to which the production line abnormality determination device 1 of the third embodiment is applied, similar to the production line abnormality determination system S to which the production line abnormality determination device 1 of the first embodiment is applied, the abnormality determination unit 1C learns the determination criteria used to determine whether or not there is an abnormality in the production line PL, and the abnormality determination unit 1C that has performed supervised learning determines whether or not there is an abnormality in the production line PL based on the production line moving image LM. Therefore, when the abnormality determination unit 1C determines that an abnormality has occurred on the production line PL, an operator or manager of the production line PL, for example, can easily identify and confirm products P included in the production line moving image LM that may have become defective by using the image capture time of the production line moving image LM, the image capture time of the identification code image DM, the time required for the product P to move between the position of the product P included in the production line moving image LM and the position of the product P included in the identification code image DM, and the identification code PD (identification code PD identified by the OCR determination unit 1F) attached to the product P included in the identification code image DM.

[0041] <Fourth embodiment> A fourth embodiment of the production line abnormality determination device, production line abnormality determination system, production line abnormality determination method, and program of the present invention will be described below. The production line abnormality determination system S to which the production line abnormality determination device 1 of the fourth embodiment is applied is configured similarly to the production line abnormality determination system S of the above-described third embodiment, except for the points described below. Therefore, the production line abnormality determination system S of the fourth embodiment can achieve the same effects as the production line abnormality determination system S of the above-described third embodiment, except for the points described below.

[0042] A production line abnormality determination system S to which the production line abnormality determination device 1 of the fourth embodiment is applied is configured similarly to the production line abnormality determination system S shown in Fig. 1. The production line abnormality determination system S of the fourth embodiment determines whether or not there is an abnormality in the production line PL (see Fig. 2).

[0043] In the fourth embodiment of the abnormality judgment device 1 for a manufacturing line, similar to the third embodiment of the abnormality judgment device 1 for a manufacturing line, the abnormality judgment unit 1C judges whether or not there is an abnormality in the manufacturing line PL (the manufacturing line PL for products P other than bag-in-box B) based on the manufacturing line moving image LM acquired by the moving image acquisition unit 1A. In detail, as described above, in the manufacturing line abnormality determination device 1 of the third embodiment, the abnormality determination unit 1C learns the determination criteria used to determine whether or not an abnormality exists in the manufacturing line PL. On the other hand, in the manufacturing line abnormality determination device 1 of the fourth embodiment, the abnormality determination unit 1C does not learn the determination criteria used to determine whether or not there is an abnormality in the manufacturing line PL.

[0044] In the fourth embodiment of the abnormality determination device 1 for a manufacturing line, when the manufacturing line moving image LM acquired by the moving image acquisition unit 1A includes an operator of the manufacturing line PL, the abnormality determination unit 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL. Even if the production line PL is a production line PL for a product P other than bag-in-box B, the operator of the production line PL is not normally within the imaging range of the imaging section S1A of the imaging device S1, and approaches the production line PL when he or she senses an abnormality or malfunction in the production line PL.

[0045] In addition, in the fourth embodiment of the abnormality determination device 1 for a manufacturing line, if the movement of the product P included in the manufacturing line moving image LM acquired by the moving image acquisition unit 1A differs from the predetermined normal movement, the abnormality determination unit 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL. Even if the product P is a product P other than a bag-in-box B, the product P usually moves at a constant speed or at a speed that changes periodically (a preset movement), and if an abnormality occurs in the production line PL, the movement of the product P will no longer be the preset movement.

[0046] Furthermore, in the fourth embodiment of the abnormality determination device 1 for a manufacturing line, if the movement of the manufacturing unit PL2 included in the manufacturing line moving image LM acquired by the moving image acquisition unit 1A differs from the predetermined normal movement (for example, if the manufacturing unit PL2 deteriorates and the operating speed of the manufacturing unit PL2 slows down), the abnormality determination unit 1C determines that an abnormality has occurred (or is likely to occur) in the manufacturing line PL. Even if the production line PL is a production line PL for a product P other than bag-in-box B, the production unit PL2 normally operates in a preset manner, and in the event of an abnormality, such as when the production unit PL2 deteriorates, the operation of the production unit PL2 will no longer be as preset.

[0047] Even in the production line abnormality judgment system S to which the production line abnormality judgment device 1 of the fourth embodiment is applied, if the abnormality judgment unit 1C judges that an abnormality has occurred in the production line PL, an operator, manager, etc. of the production line PL can easily identify and confirm the product P included in the production line moving image LM that may have become a defective product by using the imaging time of the production line moving image LM, the imaging time of the identification code image DM, the time required for the product P to move between the position of the product P included in the production line moving image LM and the position of the product P included in the identification code image DM, and the identification code PD (identification code PD identified by the OCR judgment unit 1F) attached to the product P included in the identification code image DM.

[0048] Although the present invention has been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions can be made without departing from the spirit of the present invention. The configurations described in the above-described embodiments and examples can be combined as appropriate.

[0049] Note that all or part of the functions of each unit of the production line abnormality determination device 1 in the above-described embodiment may be realized by recording a program for realizing these functions on a computer-readable recording medium, and reading and executing the program recorded on the recording medium into a computer system. Note that the term "computer system" here includes hardware such as an OS and peripheral devices. Furthermore, "computer-readable recording media" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage units such as hard disks built into computer systems. Furthermore, "computer-readable recording media" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs over networks like the Internet or communication lines like telephone lines, or devices that store programs for a fixed period of time, such as volatile memory within computer systems that serve as servers or clients in such cases. Furthermore, the above-mentioned programs may be programs that realize some of the aforementioned functions, or may be programs that can realize the aforementioned functions in combination with programs already stored in the computer system. [Explanation of symbols]

[0050] 1...abnormality determination device, 1A...video image acquisition unit, 1B...identification code image acquisition unit, 1C...abnormality determination unit, 1D...video image capture time acquisition unit, 1E...identification code image capture time acquisition unit, 1F...OCR determination unit, 1G...alarm issuance unit, S...abnormality determination system, S1...imaging device, S1A...imaging unit, S1B...recording unit, S2...imaging device, S2A...imaging unit, S2B...recording unit, PL...production line, PL1...transport unit, PL2...production unit, P...product, PD...identification code

Claims

1. a moving image acquisition unit that acquires a production line moving image that is a moving image including at least a part of a production line that manufactures products; an identification code image acquisition unit that acquires an identification code image that is an image including an identification code added to the product; an abnormality determination unit that determines whether or not there is an abnormality in the production line based on the production line moving image acquired by the moving image acquisition unit; a video image capturing time acquisition unit that acquires the video image capturing time of the production line video image; an identification code image capturing time acquiring unit that acquires the capturing time of the identification code image; When the abnormality determination unit determines that an abnormality has occurred in the production line, the identification code image acquired by the identification code image acquisition unit, the imaging time of the production line moving image acquired by the imaging time acquisition unit, and the imaging time of the identification code image acquired by the identification code image imaging time acquisition unit are used to identify the product included in the production line moving image that may have become a defective product, an OCR determination unit that identifies the identification code included in the identification code image by performing OCR (Optical Character Recognition) determination on the identification code image acquired by the identification code image acquisition unit; If the OCR determination unit can identify the identification code included in the identification code image, the identification code identified by the OCR determination unit is used to identify the product that may have become defective, If the OCR determination unit is unable to identify the identification code included in the identification code image, the identification code image is used to identify the product that may be defective. An abnormality detection device for a manufacturing line.

2. the abnormality determination unit learns a determination criterion used to determine whether or not an abnormality exists in the production line; The abnormality determination device for a manufacturing line according to claim 1.

3. The abnormality determination unit a learning production line video image that is a video image including at least a part of the production line; Information indicating whether or not an abnormality has occurred in the production line included in the learning production line video image Supervised learning is performed using training data that is a pair of information and The abnormality determination device for a manufacturing line according to claim 2.

4. When the production line moving image acquired by the moving image acquisition unit includes an operator of the production line, the abnormality determination unit determines that an abnormality has occurred in the production line. The abnormality determination device for a manufacturing line according to claim 1.

5. the production line moving image includes the product being transported by a transport unit of the production line; When the movement of the product included in the production line moving image acquired by the moving image acquisition unit differs from a predetermined normal movement, the abnormality determination unit determines that an abnormality has occurred in the production line. The abnormality determination device for a manufacturing line according to claim 1.

6. the production line video image includes a production unit that produces the product; When the movement of the manufacturing unit included in the manufacturing line moving image acquired by the moving image acquisition unit differs from a predetermined normal movement, the abnormality determination unit determines that an abnormality has occurred in the manufacturing line. The abnormality determination device for a manufacturing line according to claim 1.

7. The product is a bag-in-box having an outer box, an inner bag housed in the outer box, a spout attached to the inner bag, and a cap attached to the spout, the manufacturing line includes a conveying unit that conveys the bag-in-box and the manufacturing unit that manufactures the bag-in-box; the manufacturing unit includes at least a filling unit that fills the inner bag with contents, a capping mechanism that opens and caps the spouts, and a cutting unit that cuts the multiple inner bags that are connected to each other into individual inner bags; the conveying unit includes at least an inner bag supply section that supplies the plurality of inner bags connected to one another to the filling section, and an elevating section that elevates a filling table that supports the inner bags when the contents are filled into the inner bags by the filling section, When the movement of any of the filling unit, the inner bag supply unit, the lifting unit, the capping mechanism, and the cutting unit included in the production line moving image acquired by the moving image acquisition unit differs from a predetermined normal movement, the abnormality determination unit determines that an abnormality has occurred in the production line. The abnormality determination device for a manufacturing line according to claim 6.

8. an alarm issuing unit that issues an alarm when the abnormality determining unit determines that an abnormality has occurred in the production line; The abnormality determination device for a manufacturing line according to claim 1.

9. The abnormality determination device according to any one of claims 1 to 8; a first imaging device that captures the moving image of the production line; and a second imaging device that captures the identification code image.

10. a moving image acquisition step of acquiring a production line moving image that is a moving image including at least a part of a production line that manufactures products; an identification code image acquisition step of acquiring an identification code image which is an image including the identification code added to the product; an abnormality determination step of determining whether or not there is an abnormality in the production line based on the production line moving image acquired in the moving image acquisition step; a moving image capturing time acquisition step of acquiring a capturing time of the moving image of the production line; an identification code image capturing time acquiring step of acquiring a capturing time of the identification code image; When it is determined in the abnormality determination step that an abnormality has occurred in the production line, the identification code image acquired in the identification code image acquisition step, the imaging time of the production line moving image acquired in the imaging time acquisition step, and the imaging time of the identification code image acquired in the identification code image imaging time acquisition step are used to identify the product included in the production line moving image that may have become a defective product, an OCR (Optical Character Recognition) step of identifying the identification code included in the identification code image by performing an OCR determination on the identification code image acquired in the identification code image acquisition step; If the OCR step is able to identify the identification code included in the identification code image, the identification code identified by the OCR step is used to identify the product that may have become defective; If the OCR step is unable to identify the identification code included in the identification code image, the identification code image is used to identify the product that may be defective. A method for determining abnormalities on a production line.

11. On the computer, a moving image acquisition step of acquiring a production line moving image that is a moving image including at least a part of a production line that manufactures products; an identification code image acquisition step of acquiring an identification code image which is an image including the identification code added to the product; an abnormality determination step of determining whether or not there is an abnormality in the production line based on the production line moving image acquired in the moving image acquisition step; a moving image capturing time acquisition step of acquiring a capturing time of the moving image of the production line; a program for executing an identification code image capturing time acquisition step of acquiring an image capturing time of the identification code image, When it is determined in the abnormality determination step that an abnormality has occurred in the production line, the identification code image acquired in the identification code image acquisition step, the imaging time of the production line moving image acquired in the imaging time acquisition step, and the imaging time of the identification code image acquired in the identification code image imaging time acquisition step are used to identify the product included in the production line moving image that may have become a defective product, On the computer, executing an OCR (Optical Character Recognition) step of identifying the identification code included in the identification code image by performing an OCR determination on the identification code image acquired in the identification code image acquisition step; If the OCR step is able to identify the identification code included in the identification code image, the identification code identified by the OCR step is used to identify the product that may have become defective; If the OCR step is unable to identify the identification code included in the identification code image, the identification code image is used to identify the product that may be defective. program.

Images