JP7477301B2 - Collection device, collection method, and program for manufacturing absorbent articles - Google Patents

Description

The present invention relates to a collection device, a collection method, and a program for manufacturing absorbent articles.

Conventionally, a technology is known for manufacturing equipment that produces absorbent articles, which associates product data with equipment data, and when an abnormality occurs in a product, identifies at least one of the product data and equipment data associated with the product determined to be abnormal, and further identifies the manufacturing process that caused the product abnormality.

JP 2018-129030 A

However, the above-mentioned technology has room for improvement in terms of improving the accuracy of estimating abnormalities that may occur in manufacturing equipment that manufactures absorbent articles. For example, the above-mentioned technology may not be able to estimate in advance abnormalities that may occur in manufacturing equipment that manufactures absorbent articles.

One example of a method for predicting such anomalies in advance is to collect equipment data during production and use the equipment data as training data to generate a learning model for prediction. For this reason, when generating a learning model with high accuracy for predicting anomalies in absorbent article manufacturing equipment, it is necessary to effectively collect equipment data to be used as training data.

The present application has been made in consideration of the above, and aims to effectively collect training data for a learning model that estimates abnormalities that occur in manufacturing equipment that produces absorbent articles.

The collection device for manufacturing absorbent articles according to the present application is characterized by comprising a first collection unit that collects data from a specified sensor among a plurality of sensors provided on a manufacturing line for absorbent articles at a first speed, a second collection unit that collects data from sensors other than the specified sensor at a second speed slower than the first speed, and a transmission unit that transmits the data collected by the first collection unit and the second collection unit to a generation device that generates a learning model that uses the data as training data to estimate anomalies in the manufacturing line.

According to one aspect of the embodiment, it is possible to effectively collect training data for a learning model that estimates abnormalities that occur in manufacturing equipment that manufactures absorbent articles.

FIG. 1 is a schematic side view showing an example of the configuration of a manufacturing line according to an embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of a collection system according to the embodiment. FIG. 3 is a diagram illustrating an example of a configuration of the processing unit. FIG. 4 is an explanatory diagram of the product pitch. FIG. 5 is a block diagram illustrating an example of the configuration of a high-speed acquisition unit according to the embodiment. FIG. 6 is a block diagram showing an example of the configuration of a master collection unit included in the high-speed collection unit. FIG. 7 is a block diagram showing an example of the configuration of a slave collection unit included in the high speed collection unit. FIG. 8 is a diagram illustrating an example of a connection configuration of the master collection unit and the slave collection units. FIG. 9 is a timing chart of the high-speed collection process executed by the high-speed collection unit. FIG. 10 is a diagram showing the relationship between the phase angle indicated by the reference encoder and the sampling start timing. FIG. 11 is a block diagram illustrating an example of the configuration of a low-speed collection unit according to an embodiment. FIG. 12 is a block diagram illustrating an example of a configuration of a master collection unit included in the low-speed collection unit. FIG. 13 is a timing chart of the high-speed collection process executed by the high-speed collection unit and the low-speed collection process executed by the low-speed collection unit. FIG. 14 is a flowchart illustrating a process performed by the collection device according to the embodiment. FIG. 15 is a diagram illustrating an example of a hardware configuration.

The following points become clear at least from the description in this specification and the accompanying drawings.

A collection device for manufacturing absorbent articles, comprising: a first collection unit that collects data from a specified sensor among a plurality of sensors provided in a manufacturing line for absorbent articles at a first speed; a second collection unit that collects data from sensors other than the specified sensor at a second speed slower than the first speed; and a transmission unit that transmits the data collected by the first collection unit and the second collection unit to a generation device that generates a learning model that uses the data as training data to estimate an abnormality in the manufacturing line.

This type of collection device can effectively collect training data for a learning model that estimates abnormalities that occur in manufacturing equipment that produces absorbent articles.

In addition, in a collection device for manufacturing absorbent articles, the first collection unit and the second collection unit collect sensor data that are correlated with each other in the manufacture of the absorbent articles.

This type of collection device makes it possible to effectively collect data that takes into account the correlation between processes, for example in the manufacture of absorbent articles, which have the characteristic that the processing steps performed are often different from each other but have a high correlation between the processes.

In addition, in a collection device for manufacturing absorbent articles, the first collection unit collects at least vibration sensor data.

With such a collection device, for example, data from a vibration sensor that changes dramatically even over a short period of time can be sampled at high resolution and speed in response to such changes, and data that is useful for generating a learning model can be collected.

In addition, in a collection device for manufacturing absorbent articles, the second collection unit collects data from at least one of a pressure sensor and a temperature sensor.

With such a collection device, for example, data from pressure sensors or temperature sensors that change slowly over a short period of time can be sampled at a low speed with a resolution that corresponds to such gradual changes, and data that is useful for generating a learning model can be collected.

In addition, in a collection device for manufacturing absorbent articles, the first collection unit and the second collection unit each have a master collection unit that serves as a master and one or more slave collection units that are daisy-chained to the master collection unit, and at least the first collection unit of the first collection unit and the second collection unit performs synchronization control so that data is collected synchronously by the master collection unit and the slave collection unit.

With such a collection device, for example, it is possible to simultaneously collect data from multiple sensors placed at different positions along a production line in a synchronized manner at a specific point in time. Therefore, it is possible to collect data that indicates the correlation between sensors at a specific point in time on the production line. Furthermore, based on such data, it is possible to provide the generation device with training data that is useful for generating a learning model with high accuracy in estimating anomalies.

In addition, in a collection device for manufacturing absorbent articles, the first collection unit and the second collection unit are synchronously controlled so that data is collected by the master collection unit and the slave collection unit at the same time on the time axis.

With such a collection device, synchronization control is performed so that data is collected synchronously at the same time on the time axis, making it possible to collect data that indicates the correlation between processing steps at a specific point in time on the production line. In addition, based on such data, it is possible to provide the generation device with training data that is useful for generating a learning model with high accuracy in estimating anomalies.

The manufacturing line also has a reference device for measuring the phase angle of the manufacturing line, and one rotation of the reference device corresponds to the length of one piece of the absorbent article. In the collection device for manufacturing absorbent articles, the first collection unit and the second collection unit perform synchronization control so that data is collected by the master collection unit and the slave collection unit when the reference device indicates a specific phase angle.

With such a collection device, it is possible to collect data that indicates the correlation between processing steps at a specific point in time on the production line, synchronized with a specific phase angle of the production line, in other words, synchronized with an arbitrary position corresponding to the specific phase angle of one piece of absorbent article. In other words, it is possible to selectively set a specific position of the absorbent article, and data can be collected at any point in time based on such setting. Furthermore, based on such data, it is possible to provide the generation device with training data that is useful for generating a learning model with high accuracy in estimating anomalies.

In addition, in a collection device for manufacturing absorbent articles, the first collection unit and the second collection unit start collecting data in synchronization with each other.

Such a collection device makes it possible to collect data that indicates the correlation between different processing processes at a specific point in time on the production line, including both high-speed and low-speed collection systems. In addition, based on such data, it is possible to provide the generation device with training data that is useful for generating a learning model with high accuracy in estimating anomalies.

The manufacturing line also has multiple processing units that process the continuous product, which is the source of the absorbent article, at different positions, and in the collection device related to the manufacture of absorbent articles, the first collection unit and the second collection unit each collect data at the processing units.

This type of collection device makes it possible to collect data showing the correlation between multiple processing sections that process continuous products at different positions on the production line while maintaining a balance with each other. In addition, based on this data, it is possible to provide the generation device with training data that is useful for generating a learning model with high accuracy in estimating anomalies.

Below, an example of a form (hereinafter, referred to as an "embodiment") for implementing a collection device, collection method, and program for manufacturing absorbent articles will be described in detail with reference to the drawings. Note that the collection device, collection method, and program for manufacturing absorbent articles are not limited to this embodiment. Also, the same parts in the following embodiments are given the same reference numerals, and duplicated explanations will be omitted.

[Embodiment]
[1. Example of manufacturing line configuration]
First, prior to describing the collection device 10 according to the embodiment, a configuration example of a production line PL, which is an example of a manufacturing device for manufacturing absorbent articles, will be described with reference to Fig. 1. Fig. 1 is a schematic side view showing an example of the configuration of the production line PL according to the embodiment.

The manufacturing line PL according to the embodiment is a continuous manufacturing process for manufacturing absorbent articles. The absorbent articles are, for example, diapers, sanitary napkins, and urine pads. In the following, the main example of the manufacturing of diapers D as an absorbent article will be described.

On the production line PL, multiple processing steps are carried out at different locations on the continuous sheet (also called a "continuous web"), which is the source of diaper D. Note that "processing" here refers to all the steps that are applied to the continuous web before one piece of diaper D is finally produced.

Therefore, in addition to processes in which traces of "processing" such as sequentially arranging absorbents on a continuous web, forming the continuous web into a specified shape, and cutting it into pieces ultimately remain on one piece of diaper D, it also includes processes in which no traces of "processing" ultimately remain on one piece of diaper D, such as material splicing, which connects materials together so that the material, such as a continuous web, is not broken.

In the following, the width direction of the production line PL (the direction penetrating the paper surface of FIG. 1) will be referred to as the "CD direction", and of the two directions perpendicular to the CD direction, the vertical direction will be referred to as the "up-down direction" and the horizontal direction will be referred to as the "front-rear direction".

As shown in FIG. 1, the production line PL includes a core wrap transport path R1, an absorbent transport path R2, a fastening tape transport path R3, a top sheet transport path R4, a target tape transport path R5, a back sheet transport path R6, and a base sheet transport path R7.

Each of the transport paths R1 to R7 is provided with a transport device (not shown). The transport device is composed of a belt conveyor, transport rollers, etc. The belt conveyor is, for example, a normal belt conveyor with a transport surface of an endless belt that rotates, or a suction belt conveyor that has a suction function on the outer surface of the endless belt.

In the core wrap conveying path R1, the core wrap sheet Cs is unwound from the material coil 201 in which the core wrap sheet Cs is wound into a coil shape. That is, in the core wrap conveying path R1, the core wrap sheet Cs, which is a continuous sheet, is conveyed. The core wrap sheet Cs is, for example, a liquid-permeable sheet member such as tissue paper or nonwoven fabric.

In the absorbent transport path R2, the absorbent Ab is placed on the core wrap sheet Cs transported from the core wrap transport path R1. The absorbent Ab is placed on the core wrap sheet Cs by the fiber stacking drum 202, which rotates around a rotation axis along the CD direction. The absorbent Ab is a liquid absorbent material, such as pulp fiber and superabsorbent polymer (SAP).

On the outer peripheral surface of the fiber stacking drum 202, multiple recesses 202a are formed along the direction of rotation. Pulp fibers and SAP sprayed from the spray duct are stacked in the recesses 202a. The recesses 202a are formed so that the shape of the absorbent Ab placed on the core wrap sheet Cs is approximately rectangular in a plan view. Multiple absorbents Ab are placed side by side on the core wrap sheet Cs along the front-to-rear direction.

A cutting device 203 is also provided on the absorbent transport path R2. The cutting device 203 cuts the core wrap sheet Cs on which the absorbent Ab is placed. The cutting device 203 includes a cutter roll 203a and an anvil roll 203b.

The cutter roll 203a rotates around a rotation axis aligned with the CD direction. The cutter roll 203a is provided with a cutter blade aligned with the rotation axis direction. The anvil roll 203b rotates around a rotation axis aligned with the CD direction.

The cutting device 203 cuts the core wrap sheet Cs on which the absorbent Ab is placed by clamping it with a cutter roll 203a and anvil roll 203b. The cutting device 203 cuts the core wrap sheet Cs at a position between adjacent absorbents Ab.

In the absorbent transport path R2, the core wrap sheet Cs cut by the cutting device 203 is transported forward.

The fastening tape Ft1, which is a continuous sheet, is transported along the fastening tape transport path R3. Along the fastening tape transport path R3, adhesive is applied to the fastening tape Ft1 by the adhesive application device 204.

In the top sheet conveying path R4, the top sheet Ts is unwound from the material coil 205 in which the top sheet Ts is wound into a coil shape. In other words, in the top sheet conveying path R4, the top sheet Ts, which is a continuous sheet, is conveyed. The top sheet Ts is a sheet member that is liquid permeable, and is, for example, a nonwoven fabric containing thermoplastic resin fibers such as polyethylene or polypropylene.

A slip cut device 206 is also provided on the top sheet transport path R4. The slip cut device 206 cuts the fastening tape Ft1 transported on the fastening tape transport path R3. The slip cut device 206 includes a cutter roll 206a and an anvil roll 206b.

The cutter roll 206a rotates around a rotation axis aligned with the CD direction. The cutter roll 206a is provided with a cutter blade (not shown) that cuts the continuous sheet of fastening tape Ft1 into single-cut fastening tape Ft2. Multiple cutter blades are provided in the direction of rotation.

The anvil roll 206b adsorbs and holds the continuous fastening tape Ft1 coated with adhesive. The anvil roll 206b rotates around an axis of rotation aligned in the CD direction. The anvil roll 206b is provided with a receiving blade (not shown) that faces the cutter blade of the cutter roll 206a.

The slip cut device 206 adsorbs the continuous sheet of fastening tape Ft1 coated with adhesive using the anvil roll 206b, and cuts the continuous sheet of fastening tape Ft1 using the cutter roll 206a to produce single sheets of fastening tape Ft2.

The slip cut device 206 adsorbs the fastening tape Ft2 cut into single pieces using the anvil roll 206b and transports it to a position opposite the top sheet Ts.

In addition, a temporary press roll 207 is provided below the anvil roll 206b on the top sheet transport path R4. The temporary press roll 207 is provided to face the anvil roll 206b with the top sheet Ts sandwiched between them.

The temporary press roll 207 rotates around a rotation axis aligned with the CD direction. The temporary press roll 207 presses the fastening tape Ft2, which is adsorbed to the anvil roll 206b, toward the anvil roll 206b when the fastening tape Ft2 is transported above the top sheet Ts. As a result, the top sheet Ts, which is a continuous body, is pressed against the anvil roll 206b, and the fastening tape Ft2 is adhered to the top sheet Ts by the adhesive applied to the fastening tape Ft2. As a result, the fastening tape Ft2 is temporarily fixed to the top sheet Ts.

The main press device 208 is also provided on the top sheet transport path R4. The main press device 208 is provided downstream of the provisional press roll 207 in the transport direction of the top sheet Ts on the top sheet transport path R4.

This press device 208 permanently fixes the fastening tape Ft2 that has been temporarily fixed to the top sheet Ts. This press device 208 clamps the top sheet Ts to which the fastening tape Ft2 has been temporarily fixed between a pair of rolls, and permanently fixes the fastening tape Ft2 to the top sheet Ts.

Each roll rotates around a rotation axis along the CD direction. Of a pair of rolls, one roll reciprocates toward the other roll. In other words, the distance between the pair of rolls can be changed.

An adhesive applicator 209 is also provided on the top sheet transport path R4. The adhesive applicator 209 is provided downstream of the main press device 208 in the transport direction of the top sheet Ts. The adhesive applicator 209 applies adhesive to the top sheet Ts to which the fastening tape Ft2 is fixed. The adhesive applicator 209 applies adhesive to the non-skin side of the top sheet Ts.

The target tape Tt1, which is a continuous sheet, is transported along the target tape transport path R5. Along the target tape transport path R5, adhesive is applied to the target tape Tt1 by the adhesive application device 210.

In the backsheet conveying path R6, the backsheet Bs is unwound from a material coil 211 in which the backsheet Bs is wound into a coil shape. In other words, in the backsheet conveying path R6, the backsheet Bs, which is a continuous sheet, is conveyed. The backsheet Bs is a sheet member that is liquid-impermeable, and is, for example, a thermoplastic resin film such as polyethylene.

A slip cut device 212 is also provided on the back sheet transport path R6. The slip cut device 212 cuts the target tape Tt1 transported on the target tape transport path R5. The slip cut device 212 includes a cutter roll 212a and an anvil roll 212b.

The cutter roll 212a rotates around a rotation axis aligned with the CD direction. The cutter roll 212a is provided with a cutter blade (not shown) that cuts the continuous sheet of target tape Tt1 into single-cut target tape Tt2. Multiple cutter blades are provided in the direction of rotation.

The anvil roll 212b adsorbs and holds the continuous target tape Tt1 coated with adhesive. The anvil roll 212b rotates around an axis of rotation aligned with the CD direction. The anvil roll 212b is provided with a receiving blade (not shown) that faces the cutter blade of the cutter roll 212a.

The slip cut device 212 adsorbs the continuous sheet of target tape Tt1 coated with adhesive using the anvil roll 212b, and cuts the continuous sheet of target tape Tt1 using the cutter roll 212a to produce a single sheet of target tape Tt2.

The slip cut device 212 adsorbs the target tape Tt2 cut into single pieces using the anvil roll 212b and transports it to a position opposite the back sheet Bs.

In addition, a temporary press roll 213 is provided below the anvil roll 212b on the back sheet transport path R6. The temporary press roll 213 is provided to face the anvil roll 212b with the back sheet Bs in between.

The temporary press roll 213 rotates around a rotation axis aligned with the CD direction. The temporary press roll 213 presses the target tape Tt2, which is adsorbed to the anvil roll 212b, toward the anvil roll 212b when it is transported above the back sheet Bs. As a result, the back sheet Bs, which is a continuous body, is pressed against the anvil roll 212b, and the target tape Tt2 is adhered to the back sheet Bs by the adhesive applied to the target tape Tt2. As a result, the target tape Tt2 is temporarily fixed to the back sheet Bs.

The main press device 214 is also provided on the back sheet transport path R6. The main press device 214 is provided downstream of the temporary press roll 213 in the transport direction of the back sheet Bs on the back sheet transport path R6.

This press device 214 finally fixes the target tape Tt2 that has been temporarily fixed to the back sheet Bs. This press device 214 clamps the back sheet Bs to which the target tape Tt2 has been temporarily fixed between a pair of rolls, and finally fixes the target tape Tt2 to the back sheet Bs.

Each roll rotates around a rotation axis along the CD direction. Of a pair of rolls, one roll reciprocates toward the other roll. In other words, the distance between the pair of rolls can be changed.

An adhesive applicator 215 is also provided on the back sheet transport path R6. The adhesive applicator 215 is provided downstream of the main press device 214 in the transport direction of the back sheet Bs. The adhesive applicator 215 applies adhesive to the back sheet Bs to which the target tape Tt2 is fixed. The adhesive applicator 215 applies adhesive to the skin side of the back sheet Bs.

The absorbent Ab transported by the absorbent transport path R2, the top sheet Ts transported by the top sheet transport path R4, and the back sheet Bs transported by the back sheet transport path R6 join at a joining position Mp.

Specifically, at the joining position Mp, the back sheet Bs of the continuous sheet joins from the non-skin side of the absorbent Ab, and the top sheet Ts of the continuous sheet joins from the skin side of the absorbent Ab. Because adhesive is applied to the top sheet Ts and back sheet Bs, the top sheet Ts, absorbent Ab, and back sheet Bs are bonded and integrated by the adhesive to produce the base sheet BMs of the continuous sheet. In the base sheet BMs, the absorbent Ab is continuously lined up in the front-to-rear direction at a product pitch P equivalent to the length of one piece of diaper D.

In FIG. 1, the base sheet BMs downstream of the junction position Mp in the conveying direction of the base sheet BMs is shown in a state in which the top sheet Ts, absorbent body Ab, and back sheet Bs are separated from each other, but in reality, they are joined together.

The base sheet BMs is transported along the base sheet transport path R7. A leg hole cutting device 216 is provided on the base sheet transport path R7. The leg hole cutting device 216 cuts a portion of the base sheet BMs on both sides in the CD direction to form leg openings of the diaper D. The leg hole cutting device 216 includes a cutter roll 216a and an anvil roll 216b.

The cutter roll 216a rotates around a rotation axis aligned with the CD direction. The cutter roll 216a is provided with a cutter blade (not shown) along the rotation direction. The cutter blade is curved to match the shape of the leg opening. The anvil roll 216b rotates around a rotation axis aligned with the CD direction.

The rotation of each roll 216a, 216b in the leg hole cutting device 216 is linked to the conveying operation of the base sheet BMs so that leg openings are formed at predetermined positions in the base sheet BMs.

In the leg hole cutting device 216, the cutter roll 216a can be moved toward the anvil roll 216b, and the distance between the cutter roll 216a and the anvil roll 216b can be changed.

An end cut device 217 is also provided on the base sheet transport path R7. The end cut device 217 is provided downstream of the leg hole cut device 216 in the transport direction of the base sheet BMs on the base sheet transport path R7.

The end cut device 217 cuts the base sheet BMs transported by the base sheet transport path R7. The end cut device 217 includes a cutter roll 217a and an anvil roll 217b.

The cutter roll 217a rotates around a rotation axis aligned with the CD direction. The cutter roll 217a is provided with a cutter blade (not shown) along the rotation axis direction. The anvil roll 217b rotates around a rotation axis aligned with the CD direction.

The end cut device 217 cuts the downstream end of the base sheet BMs at a preset position on the base sheet BMs to produce the diaper D.

In this way, in the manufacturing line PL, various processing devices (hereinafter referred to as "processing section 300") involved in the production of diapers D, such as material coils 201, 205, 211, fiber stacking drum 202, cutting device 203, adhesive application devices 204, 209, 210, 215, slip cut devices 206, 212, temporary press rolls 207, 213, main press devices 208, 214, leg hole cut device 216, end cut device 217, and conveying devices that form each conveying route R1 to R7, work in conjunction with each other while maintaining a continuous web at a predetermined tension to produce diapers D.

In other words, the diaper D is manufactured through different but linked processing steps, such as conveying the continuous web at high speed while maintaining a predetermined tension, sequentially arranging the absorbents Ab, forming the web into a predetermined shape, and cutting the web into individual pieces of diaper D, by a plurality of processing sections 300 provided at different positions on the manufacturing line PL. That is, while the processing steps on the manufacturing line PL often differ in their respective aspects, there is a high correlation between the processing steps, such as downstream processing steps being easily affected by upstream processing steps.

Therefore, when attempting to estimate abnormalities in the production line PL by collecting equipment data during production on the production line PL and using a learning model generated using such equipment data as training data, it is necessary to collect data effectively, taking into account the correlation between the processing steps described above.

Therefore, in the collection method according to the embodiment, data from a predetermined sensor among a plurality of sensors provided on the absorbent article production line PL is collected at a first speed, and data from sensors other than the predetermined sensor is collected at a second speed slower than the first speed. Then, the data collected at the first speed and the data collected at the second speed are transmitted to a generation device that generates a learning model for estimating anomalies in the production line PL using the data as training data.

Below, a detailed description will be given of an example configuration of a collection system 1 that applies the collection method according to this embodiment, with reference to Figure 2 onwards.

2. Example of the configuration of the collection system according to the embodiment
Fig. 2 is a block diagram showing an example of the configuration of the collection system 1 according to the embodiment. Note that, although various block diagrams are shown in Fig. 2 as well as Fig. 5 to Fig. 7, Fig. 11, and Fig. 12 shown later, these block diagrams show only components necessary for explaining the features of this embodiment, and descriptions of general components are omitted.

In other words, each component shown in these block diagrams is a functional concept, and does not necessarily have to be physically configured as shown. For example, the specific form of distribution and integration of each block is not limited to that shown, and all or part of it can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

In addition, when using these block diagrams, the explanation of components that have already been explained may be simplified or omitted.

As shown in FIG. 2, the collection system 1 according to the embodiment includes a collection device 10, a display unit 20, a generation device 500, and a production line PL.

The collection device 10, the display unit 20, and the production line PL are communicatively connected to each other via a network 100, which is a wired or wireless communication line. The network 100 is, for example, an intranet configured by a LAN (Local Area Network) or the like.

The collection device 10 and the generation device 500 are communicatively connected to each other via a network N, which is a wired or wireless communication line. The network N is, for example, a communication network such as a LAN, a WAN (Wide Area Network), a telephone network (a mobile phone network, a landline telephone network, etc.), a regional IP (Internet Protocol) network, or the Internet.

[2-1. About the generating device]
Here, the generating device 500 will be described. The generating device 500 is a device that generates a learning model 501 that estimates an abnormality in the production line PL by using the data collected by the collecting device 10 as teacher data. The generating device 500 is realized as, for example, a cloud server. The generating device 500 generates the learning model 501 by using a predetermined machine learning algorithm.

As an algorithm for machine learning, for example, deep learning can be used, but this is not limited thereto. Machine learning may be performed by a regression analysis method such as support vector regression using a pattern classifier such as a support vector machine (SVM). In addition, the pattern classifier is not limited to an SVM, and may be, for example, adaptive boosting. In addition, random forests, etc. may also be used.

The generating device 500 also distributes the generated learning model 501 to the production line PL, for example, via the collecting device 10. The production line PL estimates anomalies in the production line PL, for example, by inputting real-time equipment data during production into the distributed learning model 501.

The generating device 500 may store the learning model 501 without distributing it, and may estimate abnormalities in the manufacturing line PL based on real-time data during manufacturing that is uploaded from the collecting device 10 via the network N at any time.

[2-2. Example of collection device configuration]
The collection device 10 includes a storage unit 11, a high-speed collection unit 12, a low-speed collection unit 13, and a transmission unit 14. The storage unit 11 is realized by a storage device such as a network attached storage (NAS), a hard disk, or an optical disk, and stores a collection DB (database) 11a in the example of Fig. 2.

The collection DB 11a is a database that accumulates data collected by the high-speed collection unit 12 and the low-speed collection unit 13.

The high-speed collection unit 12 corresponds to an example of a "first collection unit" and collects data from a specific sensor Sr among the multiple sensors Sr provided on the production line PL at a first speed and stores the collected data in the collection DB 11a.

The low-speed collection unit 13 corresponds to an example of a "second collection unit" and collects data from sensors Sr other than the sensors Sr targeted for collection by the high-speed collection unit 12 at a second speed slower than the first speed, and stores the collected data in the collection DB 11a. The high-speed collection unit 12 and the low-speed collection unit 13 collect data from sensors Sr that are correlated with each other in the manufacture of diapers D. Configuration examples of the high-speed collection unit 12 and the low-speed collection unit 13 will be described in more detail with reference to FIG. 5 and subsequent figures.

The transmission unit 14 acquires the data collected by the high-speed collection unit 12 and the low-speed collection unit 13 from the collection DB 11a and transmits it to the generation device 500 via the network N.

The display unit 20 is an information display device equipped with a display or the like, and is configured to be able to appropriately display the processing status of the collection device 10, for example, the data collection status of the high-speed collection unit 12, the data collection status of the low-speed collection unit 13, the data accumulation status of the collection DB 11a, the data transmission status of the transmission unit 14, etc.

The display unit 20 may be, for example, a mobile phone including a smartphone, a tablet terminal, a desktop PC (Personal Computer), a notebook PC, a PDA (Personal Digital Assistant), or other information processing device. The display unit 20 may also be a wearable device that is a glasses-type or watch-type information processing terminal.

The manufacturing line PL includes a plurality of processing units 300 (300-1, 300-2, 300-3, etc.) and a reference encoder RE. The plurality of processing units 300 correspond to various processing devices involved in the manufacture of diapers D, such as the material coils 201, 205, 211, the fiber stacking drum 202, the cutting device 203, the adhesive application devices 204, 209, 210, 215, the slip cut devices 206, 212, the temporary press rolls 207, 213, the main press devices 208, 214, the leg hole cut device 216, the end cut device 217, and the conveying devices that form the respective conveying paths R1 to R7.

Various sensors Sr (Sr-1, Sr-2, Sr-3, etc.) are connected to each processing unit 300. The reference encoder RE is an example of a reference device that measures the phase angle of the production line PL.

[2-3. Various sensors and reference encoders]
Here, the various sensors Sr and the reference encoder RE will be described. Fig. 3 is a diagram showing an example of the configuration of the processing unit 300. Fig. 4 is an explanatory diagram of the product pitch P. Note that Fig. 3 shows the above-mentioned cutting device 203 as an example of the processing unit 300.

As already mentioned, as shown in FIG. 3, the cutting device 203 includes a cutter roll 203a and an anvil roll 203b. The cutting device 203 also includes a motor 203c and an encoder 203d.

The motor 203c is a drive source that rotates the cutter roll 203a. The encoder 203d is provided at the shaft end of the motor 203c and measures the rotation angle of the motor 203c (i.e., the rotation angle of the cutter roll 203a).

Here, the circumference of the cutter roll 203a is set to the same value as the length of the product pitch P, which is the length of one piece of diaper D shown in FIG. 4. Therefore, when the cutter roll 203a rotates once, for example, the absorbent Ab is transported by a transport amount equal to the length of the product pitch P (hereinafter, appropriately referred to as the "unit transport amount").

The encoder 203d rotates together with the motor 203c (i.e., the cutter roll 203a) and outputs a digital value proportional to the conveyance amount, for example a value ranging from 0 to a predetermined upper limit value, corresponding to 0° to 360°, for each unit conveyance amount based on the input of this rotational motion.

Such a digital value is sent to a control device (not shown) of the production line PL as, for example, a reference signal and is used for a series of interlocking controls by the multiple processing units 300 in the production line PL. The encoder 203d that outputs such a reference signal is also called the "reference encoder RE" to distinguish it from other encoders. In this way, one rotation of the reference encoder RE corresponds to the length of one piece of diaper D, and the phase angle of the production line PL is measured in correspondence with the length of one piece.

Note that, although the case where the encoder 203d of the cutting device 203 is the reference encoder RE has been described here, an encoder provided in another processing unit 300 may also be used as the reference encoder RE.

For example, any of the encoders provided in the slip cut devices 206, 212, the provisional press rolls 207, 213, the main press devices 208, 214, the leg hole cut device 216, and the end cut device 217 may be the reference encoder RE. Also, any of the encoders provided in the conveying devices forming each of the conveying paths R2, R4, R6, R7, etc. may be the reference encoder RE. Also, the reference encoder RE does not necessarily have to be an encoder physically provided in the production line PL, and may be a virtual encoder.

Furthermore, as shown in FIG. 3, various sensors Sr, such as at least a vibration sensor Sr_A, a pressure sensor Sr_P, and a temperature sensor Sr_T, are connected to the cutting device 203.

The vibration sensor Sr_A is, for example, an acceleration sensor, and measures the vibration of the cutting device 203 during the production of the diaper D. The pressure sensor Sr_P measures the pressure of the cutting device 203 during the production of the diaper D, for example, the pressure when the core wrap sheet Cs on which the absorbent Ab is placed is clamped and cut. The temperature sensor Sr_T measures the temperature of the cutting device 203 during the production of the diaper D.

[3. Example of high-speed acquisition unit configuration]
Next, an example of the configuration of the high-speed acquisition unit 12 according to the embodiment will be described with reference to Fig. 5 to Fig. 10. First, Fig. 5 is a block diagram showing an example of the configuration of the high-speed acquisition unit 12 according to the embodiment.

As shown in FIG. 5, the high-speed collection unit 12 includes a master collection unit 121 and one or more slave collection units 122 (122-1, 122-2, 122-3, etc.).

The master collection unit 121 and the slave collection unit 122 are each realized, for example, by a PLC (Programmable Logic Controller). The master collection unit 121 comprehensively controls the high-speed collection process executed by the high-speed collection unit 12.

The slave collection unit 122 samples data from, for example, the vibration sensor Sr_A among the various sensors Sr connected to each of the multiple processing units 300. Note that while FIG. 5 shows an example in which the slave collection unit 122 and the vibration sensor Sr_A are in one-to-one correspondence, the slave collection unit 122 and the vibration sensor Sr_A may also be in one-to-many correspondence.

At a predetermined collection timing, the master collection unit 121 notifies each of the slave collection units 122 of a data sampling trigger, causing all slave collection units 122 to perform high-speed sampling of the data from the vibration sensor Sr_A in synchronization with each other at the same time on the time axis.

The high-speed sampling referred to here is performed when the rotation speed of a specific motor on the production line PL during the production of diapers D is equal to or greater than a specific value, for example, under conditions of 2 seconds every 30 minutes, a sampling period of 80 microseconds, and 22,500 sampling points.

The master collection unit 121 also causes each of the slave collection units 122 to accumulate the high-speed sampled data from the vibration sensor Sr_A in the collection DB 11a.

This allows data from each vibration sensor Sr_A, which changes drastically even over a short period of time, to be sampled at high resolution and speed in response to such changes, and allows data that is useful for generating the learning model 501 to be collected.

In FIG. 5, only the slave collection unit 122 collects data from the vibration sensor Sr_A, but the master collection unit 121 may also sample data from the vibration sensor Sr_A associated with the master collection unit 121 together with the slave collection unit 122.

The following describes in more detail an example of the configuration of the master collection unit 121 and the slave collection unit 122. FIG. 6 is a block diagram showing an example of the configuration of the master collection unit 121 provided in the high-speed collection unit 12. FIG. 7 is a block diagram showing an example of the configuration of the slave collection unit 122 provided in the high-speed collection unit 12.

As shown in FIG. 6, the master collection unit 121 includes a control unit 121a. The control unit 121a is a controller, and is realized, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs stored in a storage device inside the master collection unit 121 using a RAM (Random Access Memory) as a working area. The control unit 121a can also be realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

The control unit 121a has an acquisition unit 121aa, a generation unit 121ab, and a notification unit 121ac, and realizes or executes the functions and actions of the information processing described below.

The acquisition unit 121aa acquires a predetermined collection timing. The acquisition unit 121aa acquires the predetermined collection timing based on a clock signal output from a clock generation circuit (not shown) provided in the collection device 10, for example.

Note that the collection timing referred to here is, strictly speaking, the trigger generation timing for the master collection unit 121. In addition, the acquisition unit 121aa acquires a reference signal indicating the phase angle of the production line PL from the reference encoder RE.

When the acquisition unit 121aa acquires the aforementioned collection timing, the generation unit 121ab generates a data sampling trigger for each of the slave collection units 122. The trigger includes, for example, the sampling conditions described above. The notification unit 121ac notifies each of the slave collection units 122 of the trigger generated by the generation unit 121ab.

Next, as shown in FIG. 7, the slave collection unit 122 includes a control unit 122a. The control unit 122a is a controller like the control unit 121a, and is realized, for example, by a CPU, an MPU, or the like, executing various programs stored in a storage device inside the slave collection unit 122 using a RAM as a working area. The control unit 122a can also be realized, for example, by an integrated circuit such as an ASIC or an FPGA.

The control unit 122a has an acquisition unit 122aa and a collection unit 122ab, and realizes or executes the information processing functions and actions described below.

The acquisition unit 122aa acquires a trigger notified from the master collection unit 121. The collection unit 122ab performs high-speed sampling of the data of the vibration sensor Sr_A based on the trigger acquired by the acquisition unit 122aa.

Here, based on the explanation so far, the high-speed collection process executed by the high-speed collection unit 12 will be explained in more detail. Figure 8 is a diagram showing an example of the connection configuration of the master collection unit 121 and the slave collection unit 122. Also, Figure 9 is a timing chart of the high-speed collection process executed by the high-speed collection unit 12.

As shown in FIG. 8, one or more slave collection units 122 are daisy-chained to the master collection unit 121, for example, by direct lines. The master collection unit 121 knows the communication speed using the direct lines and the distances d1, d2, d3, ... of each direct line, and can cause a trigger to reach each of the slave collection units 122 at any timing.

9, three slave collection units 122-1, 122-2, and 122-3 are connected to the master collection unit 121, and the master collection unit 121 generates a trigger and sends a notification at time t _n-4 . Then, for example, the slave collection unit 122-1 acquires a trigger at time t _n-3 . However, the master collection unit 121 does not cause the slave collection unit 122-1 to start sampling data at this time t _n-3 .

Similarly, for example, the slave collection unit 122-2 acquires a trigger at time t _n-2 , but the master collection unit 121 does not cause the slave collection unit 122-2 to start sampling data at this time t _n-2 .

For example, the slave collection unit 122-3 acquires a trigger at time t _n-1 . In response to the trigger, the master collection unit 121 causes the slave collection units 122 to simultaneously start sampling data at time t _n , which is approximately the same as time t _n-1 .

In this way, by performing synchronization control so that data is collected synchronously at the same time on the time axis, data showing the correlation between the processing units 300 at a specific point in time on the production line PL can be collected. Based on this, it is possible to provide the generation device 500 with teacher data that is useful for generating a learning model 501 with high accuracy in estimating anomalies.

Note that here, synchronization control is performed so that data is collected synchronously at the same time on the time axis, but synchronization control may also be performed that takes into account the phase angle of the reference encoder RE.

10 is a diagram showing the relationship between the phase angle indicated by the reference encoder RE and the sampling start timing. For example, it is assumed that the initial sampling start timing is time _tn as described with reference to FIG.

Then, for example, the master collection unit 121 can perform synchronization control based on the time t _n and the reference signal obtained from the reference encoder RE so that data is collected when the production line PL indicates a specific phase angle.

For example, as shown in FIG. 10, the master collection unit 121 causes the slave collection units 122 to simultaneously start sampling data at time tn ₊₁ , which is the time after time _tn when the phase angle of the production line PL first returns from 360° to the origin position of 0°.

This makes it possible to collect data showing the correlation between processing units 300 at a specific point in time on the production line PL, synchronized with a predetermined position for one piece of diaper D. Based on this, it is also possible to provide the generation device 500 with teacher data that is useful for generating a learning model 501 with high accuracy in estimating anomalies.

Note that in FIG. 10, an example is given in which sampling is started when the phase angle of the production line PL indicates the origin position, but this is not limited to the origin position and may be any phase angle position.

4. Example of the low-speed collection unit configuration
Next, a configuration example of the low-speed collection unit 13 according to the embodiment will be described with reference to Fig. 11 and Fig. 12. First, Fig. 11 is a block diagram showing an example of the configuration of the low-speed collection unit 13 according to the embodiment.

As shown in FIG. 11, the low-speed collection unit 13 includes a master collection unit 131 and one or more slave collection units 132 (132-1, 132-2, 132-3, etc.).

The master collection unit 131 and the slave collection unit 132 can be realized, for example, by a PLC, similar to the master collection unit 121 and the slave collection unit 122 of the high-speed collection unit 12 described above. However, as another example, the master collection unit 131 is realized by a PLC, and the slave collection unit 132 is realized by a dedicated device specialized for sampling data.

The master collection unit 131 performs overall control of the low-speed collection process executed by the low-speed collection unit 13. The slave collection unit 132 samples data from, for example, the pressure sensor Sr_P among the various sensors Sr connected to each of the multiple processing units 300. Note that, although the slave collection unit 132 and the pressure sensor Sr_P are in a 1:1 correspondence in FIG. 11, the slave collection unit 132 and the pressure sensor Sr_P may also be in a 1:n correspondence (n is a natural number).

At a predetermined collection timing, the master collection unit 131 remotely controls each of the slave collection units 132 to perform slow sampling of the data from the pressure sensor Sr_P in synchronization with the slave collection units 132 at the same time on the time axis.

The low-speed sampling referred to here is performed when the rotation speed of a specific motor on the production line PL during the production of diapers D is equal to or higher than a specific value, for example, for 2 seconds every 30 minutes, with a sampling period of 5 milliseconds and 400 sampling points.

The master collection unit 131 also causes each of the slave collection units 132 to accumulate the slow-sampled data of the pressure sensor Sr_P in the collection DB 11a.

This allows data from each pressure sensor Sr_P, which changes slowly over a short period of time, to be sampled at a resolution that corresponds to the slow changes, and data that is useful for generating the learning model 501 can be collected.

In FIG. 11, only the slave collection unit 132 collects data from the pressure sensor Sr_P, but the master collection unit 131 may also sample data from the pressure sensor Sr_P associated with the master collection unit 131 together with the slave collection unit 132.

In addition, while FIG. 11 uses a pressure sensor Sr_P as an example, the sensor Sr that measures data that changes relatively slowly over a specified period of time may be the subject of low-speed collection, and therefore may be a temperature sensor Sr_T.

A configuration example of the master collection unit 131 will be described in more detail. Figure 12 is a block diagram showing an example of the configuration of the master collection unit 131 provided in the low-speed collection unit 13.

As shown in FIG. 12, the master collection unit 131 includes a control unit 131a. The control unit 131a is a controller, similar to the above-mentioned control units 121a and 122a, and is realized, for example, by a CPU, an MPU, or the like, executing various programs stored in a storage device inside the master collection unit 131 using a RAM as a working area. The control unit 131a can also be realized, for example, by an integrated circuit such as an ASIC or an FPGA.

The control unit 131a has an acquisition unit 131aa and a remote control unit 131ab, and realizes or executes the information processing functions and actions described below.

The acquisition unit 131aa acquires a predetermined collection timing. The acquisition unit 131aa acquires the predetermined collection timing based on a clock signal output from a clock generation circuit (not shown) provided in the collection device 10, for example. The acquisition unit 131aa also acquires a reference signal indicating the phase angle of the production line PL from the reference encoder RE.

When the acquisition unit 131aa acquires the aforementioned collection timing, the remote control unit 131ab remotely controls each of the slave collection units 132, which are dedicated devices specialized for data sampling, and causes the slave collection units 132 to simultaneously start sampling data.

In addition, at this time, the remote control unit 131ab can cause the slave collection units 132 to simultaneously start sampling data when the production line PL indicates a specific phase angle based on the reference signal acquired by the acquisition unit 131aa, as described with reference to FIG. 10.

This makes it possible to collect data showing the correlation between processing units 300 at a specific point in time on the production line PL, synchronized with a predetermined position for one piece of diaper D. Based on this, it is also possible to provide the generation device 500 with teacher data that is useful for generating a learning model 501 with high accuracy in estimating anomalies.

In the explanation with reference to Figures 11 and 12, it has been described that the low-speed collection unit 13 performs synchronous control of data collection in the same manner as the high-speed collection unit 12, but the low-speed collection unit 13 does not necessarily have to perform such synchronous control.

In addition, in the explanation with reference to Figures 11 and 12, it has been stated that the master collection unit 131 of the low-speed collection unit 13 is realized by a PLC, and the slave collection unit 132 is realized by a dedicated device specialized for sampling data, but this is not limited to the above. For example, it is also possible for both the master collection unit 131 and the slave collection unit 132 to be realized by a PLC, and for the low-speed collection unit 13 to have the same configuration as the high-speed collection unit 12 shown in Figures 5 to 10.

Up to this point, we have explained the synchronous control of data collection within the high-speed collection unit 12 and the low-speed collection unit 13, but it is preferable for the high-speed collection unit 12 and the low-speed collection unit 13 to start collecting data in synchronization with each other. This point will be explained next. Figure 13 is a timing chart of the high-speed collection process performed by the high-speed collection unit 12 and the low-speed collection process performed by the low-speed collection unit 13.

As shown in FIG. 13, the high-speed collection unit 12 and the low-speed collection unit 13 each start collecting data in synchronization with each other.

The high-speed collection unit 12 and the low-speed collection unit 13 acquire a predetermined collection timing based on a clock signal output from a clock generation circuit (not shown) included in the collection device 10. Based on the collection timing, the high-speed collection unit 12 and the low-speed collection unit 13 perform control to synchronize with each other so that data collection is started at the same timing (times t _n , t _n+m ... in the example of FIG. 13).

Note that, for example, taking into consideration delays due to processing load, etc., timing may be synchronized between the high-speed collection unit 12 and the low-speed collection unit 13 by exchanging simple packets equivalent to ACK (Acknowledgement)/NAK (Negative-Acknowledgement) before actually starting to collect data. Timing may also be synchronized by using either one of the high-speed collection unit 12 or the low-speed collection unit 13 as a reference and correcting the timing of the other.

In this way, the high-speed collection unit 12 and the low-speed collection unit 13 start collecting data in synchronization with each other, making it possible to collect data that indicates the correlation between the processing units 300 at a specific point in time on the production line PL, including both the high-speed collection system and the low-speed collection system. Based on this, it is possible to provide the generation device 500 with teacher data that is useful for generating a learning model 501 with high accuracy in estimating anomalies.

5. Processing Procedure
Next, a process procedure executed by the collection device 10 according to the embodiment will be described with reference to Fig. 14. Fig. 14 is a flowchart showing a process procedure executed by the collection device 10 according to the embodiment.

First, the high-speed collection unit 12 and the low-speed collection unit 13 determine whether or not it is collection time (step S101). If it is not collection time (step S101, No), the process from step S101 is repeated.

If it is collection time (step S101, Yes), the high-speed collection unit 12 collects data from the vibration sensor Sr_A at a first speed (step S102). At the same time, the low-speed collection unit 13 collects data from sensors other than the vibration sensor Sr_A at a second speed slower than the first speed (step S103).

Then, the high-speed collection unit 12 and the low-speed collection unit 13 store the data they have collected in the collection DB 11a (step S104).

Then, the transmission unit 14 transmits the data collected by the high-speed collection unit 12 and the low-speed collection unit 13 and stored in the collection DB 11a to the generation device 500 of the learning model 501 that estimates an abnormality in the production line PL (step S105). The collection device 10 then repeats the process from step S101.

Up to this point, the various sensors Sr have been exemplified by vibration sensors Sr_A, pressure sensors Sr_P, and temperature sensors Sr_T, but the types of sensors are not limited. Therefore, any sensor that is provided on the absorbent article production line PL may be used, such as a tension sensor that measures the tension of a continuous web. In the case of a tension sensor, the change in the measured value is considered to be gradual, so it may be treated as a low-speed collection system.

[6. Other]
Among the above-mentioned processes, all or part of the processes described as being performed automatically may be performed manually. Also, all or part of the processes described as being performed manually may be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings may be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

The components of each device shown in the figure are conceptual and functional, and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure. Furthermore, each component may be functionally or physically distributed and integrated in any unit, in whole or in part, depending on various loads and usage conditions. Furthermore, each of the processes described above may be executed in appropriate combinations as long as there are no contradictions.

7. Hardware Configuration
The collection device 10 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in Fig. 15, for example. Fig. 15 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to an output device 1010 and an input device 1020, and is connected to a calculation device 1030, a cache 1040 which is a primary storage device, a memory 1050 which is a secondary storage device, an output IF (Interface) 1060, an input IF 1070, and a network IF 1080 by a bus 1090.

The arithmetic device 1030 operates based on programs stored in the cache 1040 or memory 1050, or programs read from the input device 1020, and executes various processes. The cache 1040 is a cache such as a RAM that temporarily stores data used by the arithmetic device 1030 for various calculations. The memory 1050 is a storage device in which data used by the arithmetic device 1030 for various calculations and various databases are registered, and is a memory realized by a ROM (Read Only Memory), a HDD (Hard Disk Drive), a flash memory, or the like.

The output IF 1060 is an interface for transmitting information to be output to an output device 1010 such as a monitor or printer that outputs various types of information, and may be realized by a connector conforming to a standard such as USB (Universal Serial Bus), DVI (Digital Visual Interface), or HDMI (registered trademark) (High Definition Multimedia Interface). On the other hand, the input IF 1070 is an interface for receiving information from various input devices 1020 such as a mouse, keyboard, scanner, etc., and may be realized by a USB, etc.

For example, the input device 1020 may be realized by a device that reads information from an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. The input device 1020 may also be realized by an external storage medium such as a USB memory.

The network IF 1080 has the function of receiving data from other devices via the network N and sending it to the computing device 1030, and also transmitting data generated by the computing device 1030 to other devices via the network N.

Here, the arithmetic device 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic device 1030 loads a program from the input device 1020 or the memory 1050 onto the cache 1040 and executes the loaded program. For example, when the computer 1000 functions as the collection device 10, the arithmetic device 1030 of the computer 1000 realizes the functions of the high-speed collection unit 12, the low-speed collection unit 13, and the transmission unit 14 by executing the program loaded onto the cache 1040.

The above describes the embodiments of the present application in detail with reference to the drawings. However, these are merely examples, and the embodiments of the present application can be implemented in other forms that incorporate various modifications and improvements based on the knowledge of those skilled in the art, including the forms described in the Disclosure of the Invention section. In addition, the above-mentioned "section, module, unit" can be read as "means" or "circuit," etc.

REFERENCE SIGNS LIST 1 Collection system 10 Collection device 12 High-speed collection section 13 Low-speed collection section 14 Transmission section 121 Master collection section 122 Slave collection section 131 Master collection section 132 Slave collection section 300 Processing section 500 Generation device 501 Learning model D Diaper PL Production line RE Reference encoder Sr Various sensors Sr_A Vibration sensor Sr_P Pressure sensor Sr_T Temperature sensor

Claims (11)

a first collecting unit having a master collecting unit and a slave collecting unit connected to the master collecting unit , and collecting data from a predetermined sensor among a plurality of sensors provided in a production line of absorbent articles at a first speed;
a second collection unit having a master collection unit serving as a master and a slave collection unit connected to the master collection unit, and collecting data from sensors other than the predetermined sensor at a second speed slower than the first speed;
a transmission unit that transmits the data collected by the first collection unit and the second collection unit to a generation device that generates a learning model that estimates abnormalities in the production line using the data collected by the first collection unit and the second collection unit as teacher data , wherein at least the first collection unit of the first collection unit and the second collection unit performs synchronization control so that data is collected synchronously by the master collection unit and the slave collection unit, and The first collecting unit and the second collecting unit include:
The collection device for manufacturing absorbent articles according to claim 1, further comprising: collecting sensor data correlated with each other in the manufacture of the absorbent articles. The collecting device for manufacturing absorbent articles according to claim 1 or 2, wherein the first collecting unit collects data from at least a vibration sensor. The collecting device for manufacturing absorbent articles according to claim 1 , 2 or 3 , wherein the second collecting unit collects data from at least one of a pressure sensor and a temperature sensor. The first collecting unit and the second collecting unit each include:
The system includes a master collection unit and one or more slave collection units daisy-chained to the master collection unit.
A collecting device for the production of absorbent articles according to any one of claims 1 to 4. The first collecting unit and the second collecting unit include:
The collecting device for manufacturing absorbent articles according to claim 5, characterized in that synchronous control is performed so that data is collected by the master collecting unit and the slave collecting unit at the same time on a time axis. the production line has a reference device for measuring a phase angle of the production line;
One rotation of the reference device corresponds to the length of one piece of the absorbent article,
The first collecting unit and the second collecting unit include:
The collecting device for manufacturing absorbent articles according to claim 5 or 6, characterized in that synchronous control is performed so that data is collected by the master collecting unit and the slave collecting unit when the reference device indicates a specific phase angle. The collecting device for manufacturing absorbent articles according to any one of claims 1 to 7, characterized in that the first collecting unit and the second collecting unit start collecting data in synchronization with each other. The manufacturing line has a plurality of processing units that process a continuous body, which is a source of the absorbent article, at different positions;
The collecting device for manufacturing absorbent articles according to any one of claims 1 to 8, wherein the first collecting unit and the second collecting unit each collect data in the processing unit. 1. A computer-implemented collection method comprising:
a first collection step of collecting data from a predetermined sensor among a plurality of sensors provided in a production line of absorbent articles at a first speed, the first collection step including a master collection unit and a slave collection unit connected to the master collection unit;
a second collection step having a master collection unit serving as a master and a slave collection unit connected to the master collection unit, and collecting data from sensors other than the predetermined sensor at a second speed slower than the first speed;
a transmission process for transmitting the data collected in the first collection process and the second collection process to a generation device that generates a learning model that estimates abnormalities in the production line using the data collected in the first collection process and the second collection process as teacher data . a first collection procedure including a master collection unit and a slave collection unit connected to the master collection unit, and collecting data from a predetermined sensor among a plurality of sensors provided in a production line for absorbent articles at a first speed;
a second collection procedure including a master collection unit serving as a master and a slave collection unit connected to the master collection unit, and collecting data from sensors other than the predetermined sensor at a second speed slower than the first speed;
a transmission step of performing synchronization control such that at least the first collection step of the first collection step and the second collection step is to collect data synchronously by the master collection unit and the slave collection unit , and transmitting the data collected in the first collection step and the second collection step to a generation device that generates a learning model for estimating an abnormality in the production line using the data as training data;

A program characterized by causing a computer to execute the above.

Images