JP7369993B2 - Management systems and methods, learning systems and methods, and programs - Google Patents

Description

The present disclosure generally relates to a management system and method, a learning system and method, and a program. More specifically, the present disclosure relates to a management system and method for managing articles, a learning system and method, and a program.

Patent Document 1 discloses an article management system. This article management system includes an IC tag, an IC tag reader, a positioning means for measuring the position of the IC tag reader, and a position management server. The IC tag reader transmits a read signal to the IC tag and receives a response signal from the IC tag. The location management server manages the location of the managed article to which the IC tag is attached based on the location signal obtained by the positioning means when the content of the response signal matches the content registered in the server itself.

JP2002-046821A

The present disclosure aims to provide a management system and a management method that make it easy to manage the location of items.

A management system according to one aspect of the present disclosure includes an acquisition unit, an estimation unit, an input reception unit, and a display control unit . The acquisition unit acquires article information. The article information is information about the article that is read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading section. The estimating section estimates transportation information regarding transportation of the article in a predetermined area based on read information obtained by acquiring the article information with the acquisition section. The input accepting unit accepts an input for evaluating the estimation result of the estimating unit. The display control section causes the display section to display a combination of the article information acquired by the acquisition section and the transportation information estimated by the estimation section. The transportation information includes estimated information on whether the article is to be carried in or out of the predetermined area. The reading information includes at least one of the number of times the article tag is read by the reading unit, a time interval, a frequency, the intensity of the received radio waves, and the phase of the received radio waves. There are a plurality of said articles. The display control unit displays the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the article is a target for import or export in the predetermined area. to be displayed on the display section.

A management method according to one aspect of the present disclosure is a management method executed by a computer system. The management method includes an acquisition step, an estimation step, an input reception step, and a display control step . The acquisition step is a step of acquiring article information. The article information is information about the article that is read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading section. The estimation step is a step of estimating transportation information regarding transportation of the article in a predetermined area based on the read information obtained by acquiring the article information in the acquisition step. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step. The display control step is a step of displaying a combination of the article information acquired in the acquisition step and the transportation information estimated in the estimation step on the display unit. The transportation information includes estimated information on whether the article is to be carried in or out of the predetermined area. The reading information includes at least one of the number of times the article tag is read by the reading unit, a time interval, a frequency, the intensity of the received radio waves, and the phase of the received radio waves. There are a plurality of said articles. The display control step includes displaying the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the plurality of articles are objects of import or export in the predetermined area. to be displayed on the display section.

A learning system according to one aspect of the present disclosure includes an acquisition unit, an estimation unit, an input reception unit, a display control unit, and a learning unit. The acquisition unit acquires article information. The article information is information about the article that is read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading section. The estimating unit receives as input the read information obtained by acquiring the article information by the acquiring unit and estimates transportation information regarding transportation of the article in a predetermined area using a classifier. The input accepting unit accepts an input for evaluating the estimation result of the estimating unit. The display control section causes the display section to display a combination of the article information acquired by the acquisition section and the transportation information estimated by the estimation section. The learning unit causes the classifier to perform machine learning using the information input at the input receiving unit as training data. The transportation information includes estimated information on whether the article is to be carried in or out of the predetermined area. The reading information includes at least one of the number of times the article tag is read by the reading unit, a time interval, a frequency, the intensity of the received radio waves, and the phase of the received radio waves. There are a plurality of said articles. The display control unit displays the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the article is a target for import or export in the predetermined area. to be displayed on the display section.

A learning method according to one aspect of the present disclosure is a learning method executed by a computer system. The learning method includes an acquisition step, an estimation step, an input reception step, a display control step, and a learning step. The acquisition step is a step of acquiring article information. The article information is information about the article that is read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading section. The estimating step is a step of estimating transportation information regarding transportation of the article in a predetermined area using a classifier using the read information obtained by acquiring the article information in the acquiring step as input. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step. The display control step is a step of displaying a combination of the article information acquired in the acquisition step and the transportation information estimated in the estimation step on the display unit. The learning step is a step of causing the classifier to perform machine learning using the information input in the input receiving step as training data. The transportation information includes estimated information on whether the article is to be carried in or out of the predetermined area. The reading information includes at least one of the number of times the article tag is read by the reading unit, a time interval, a frequency, the intensity of the received radio waves, and the phase of the received radio waves. There is a plurality of articles, and the display control step displays the article information and the article information for each of the plurality of articles depending on the estimated probability that the article is a target of import or export in the predetermined area. The estimated information is displayed on the display section in a predetermined positional relationship.

A program according to one aspect of the present disclosure causes one or more processors to execute the above management method or the above learning method.

The present disclosure has the advantage that the location of articles can be easily managed.

FIG. 1 is a block diagram showing the configuration of a management system and a reading system according to an embodiment of the present disclosure. FIG. 2 is a schematic diagram of the area targeted by the above management system. FIG. 3 is an explanatory diagram of the operation of the estimator in the management system mentioned above. FIG. 4 is a schematic diagram of a screen displayed on the display unit in the management system mentioned above. FIG. 5 is a schematic diagram showing an example of a screen displayed on a display unit of a terminal in the above management system. FIG. 6 is a schematic diagram showing an example of a screen displayed on a display unit of a terminal in the above management system. FIG. 7 is a schematic diagram showing an example of a screen displayed on a display unit of a terminal in the above management system. FIG. 8 is a flowchart showing the operation in the learning phase in the management system mentioned above. FIG. 9 is a flowchart showing the operation in the inference phase in the management system.

(1) Overview The management system 100 according to the present embodiment is a system used to manage information regarding the location of articles 1 in a predetermined area A1, as shown in FIGS. 1 and 2. Hereinafter, the "predetermined area A1" will be simply referred to as "area A1" unless otherwise specified. The term "article" as used in the present disclosure refers to the article 1 to be transported by the person concerned 2. Here, the number of articles 1 is not limited to one, but may be plural. In the present disclosure, the "person concerned" refers to a person who is permitted to take the article 1 out of the area A1 or bring the article 1 into the area A1, and who actually transports the article 1; This may include a person accompanying the person transporting the item 1. In other words, "transported by a person concerned" may include not only actually transporting the article 1 but also accompanying the person transporting the article 1. Here, the number of related parties 2 is not limited to one person, but may be multiple persons.

The management system 100 includes an acquisition section 5, an estimation section 31, and an input reception section 34, as shown in FIG. In this embodiment, the communication section 5, which will be described later, also serves as the acquisition section 5.

The acquisition unit 5 acquires article information. The article information is read from the article tag 11 by the reading unit 102 performing wireless communication with the article tag 11 attached to the article 1 using radio waves as a medium. Here, the reading unit 102 is installed in the area A1, and performs a process of reading the article information of the article 1 by performing wireless communication using radio waves as a medium with the article tag 11 in the area A1. That is, the reading unit 102 does not read the article information directly from the article 1, but reads the article information from the article tag 11 attached to the article 1 in a non-contact manner.

"Article information" in the present disclosure is information for identifying the article 1, and is, for example, an identification code such as a JAN (Japanese Article Number) code used in Japan. In addition to the JAN code, this type of identification code includes EPC (Electronic Product Code), EAN (European Article Number) code used in Europe, etc., and UPC (Universal Product Code) used in the United States etc. There is. In addition to identification codes such as GTIN (Global Trade Item Number) for individually identifying goods 1, EPC includes identification codes such as GRAI (Global Returnable Asset Identifier) for managing assets (goods 1) owned by companies. may include an identification code. Further, the article information is not limited to information that identifies the type (type) of the article 1, but may include information such as serial information that individually identifies articles 1 of the same type. Thereby, even if there are a plurality of articles 1 of the same type, each of the plurality of articles 1 of the same type can be specified using the article information.

The estimation unit 31 estimates transportation information regarding transportation of the article 1 in the area A1 based on the read information obtained by acquiring the article information in the acquisition section 5. In this embodiment, the transportation information is estimated by taking into consideration the related party information acquired by the acquisition unit 5. The estimating unit 31 will be described in detail in "(2.2) Management System" below.

The "reading information" in the present disclosure includes, for example, at least one of the number of times the article tag 11 is read by the reading unit 102, the time interval, the frequency, the intensity of the received radio waves, and the phase of the received radio waves. may be included. Further, "related party information" in the present disclosure is information for identifying the related party 2, and may include a unique identification code assigned to each related party 2. For example, the related party information includes at least one of the number of times, time interval, frequency, intensity of received radio waves, and phase of received radio waves of the related party tag 21 (described later) by the reading unit 102. may be included.

The “number of readings” in the present disclosure means the number of times the article information (or person information) is read by the reading unit 102 within a certain period, and more specifically, by the reading unit 102. This is the number of times article information (or related person information) was successfully read. The “reading time interval” in the present disclosure means a time interval at which article information (or stakeholder information) is read by the reading unit 102 within a certain period. More specifically, when the reading unit 102 successfully reads the article information (or stakeholder information) twice, the second article information (or stakeholder information) is read from the first reading of the article information (or stakeholder information). This is the time up to the point in time when the relevant person information) is read. The “reading frequency” in the present disclosure means the frequency at which article information (or related person information) is read by the reading unit 102 within a certain period, and more specifically, the reading unit 102 This is the frequency at which article information (or related party information) was successfully read. The “strength of received radio waves” in the present disclosure refers to the intensity of radio waves transmitted from the article tag 11 received by the antenna 101 of the reading system 10 described later or the stakeholder tag 21 described later. Signal Strength Indication) value. The “phase of received radio waves” in the present disclosure is the phase of radio waves received by the antenna 101 of the reading system 10 and transmitted from the article tag 11 or the related person tag 21.

"Transportation information" in the present disclosure may include estimated information on whether the article 1 is to be carried in or carried out in area A1. In other words, the "transportation information" is whether the article 1 is the article 1 taken out from the area A1 by the person concerned 2, or whether the article 1 is the article 1 brought into the area A1 by the person concerned 2; It may include estimated information that estimates . In addition, the "transportation information" may include the number of article information (that is, the number of articles 1) read by the reading unit 102, regardless of whether the articles are carried out from the area A1 or carried into the area A1.

The input accepting unit 34 accepts input for evaluating the estimation result of the estimating unit 31. The input is performed, for example, on a terminal 6 (for example, a personal computer or a mobile terminal such as a smartphone) owned by the person concerned 2, or a display (for example, a touch panel display) installed at a predetermined location. The “input for evaluating the estimation result” in the present disclosure is an input for evaluating whether the transportation information estimated by the estimation unit 31 is correct. As an example, when the estimation unit 31 estimates that the article 1 will be carried out from the area A1, the person concerned 2 inputs that the estimation result is correct if the article 1 is about to be carried out from the area A1. On the other hand, if the person concerned 2 is not attempting to carry out the article 1 from the area A1 or is attempting to carry the article 1 into the area A1, the person concerned 2 performs an input to correct the estimation result. In this case, the person concerned 2 inputs that the article 1 is neither to be carried in nor to be carried out, or that the article 1 is an article to be carried into the area A1.

As described above, in this embodiment, the estimation unit 31 estimates transportation information regarding transportation of the article 1. In the present embodiment, the estimation result of the estimating section 31 can be evaluated by receiving input from the person concerned 2 at the input receiving section 34. Therefore, this embodiment has the advantage that even if there is an error in the estimation result of the transportation information by the estimation unit 31, it can be corrected, and as a result, the location of the article 1 can be easily managed.

(2) Details Hereinafter, the management system 100 of this embodiment and the reading system 10 used together with the management system 100 will be described in detail. The management system 100 is introduced into, for example, an office (including a shared office), a factory, a train station, a school, a retail store such as a convenience store or a supermarket, a welfare facility, a hospital, or a facility such as a gymnasium. Furthermore, the management system 100 is not limited to being installed indoors, but may be installed outdoors. The management system 100 is used in any of the above-mentioned facilities to manage information regarding the location of the article 1, with the entire facility or a part of the facility (such as a room) as a predetermined area A1. In this embodiment, it is assumed that there is one predetermined area A1, but there may be a plurality of predetermined areas A1. In this case, the reading system 10 will be installed in each predetermined area A1.

In this embodiment, as an example, it is assumed that the management system 100 is introduced into an office. Further, in this embodiment, as shown in FIG. 2, it is assumed that the inside of the office is a predetermined area A1, and the area outside the office is an area A2. Furthermore, in this embodiment, it is assumed that a gate (for example, an automatic door) G1 is installed at the boundary between area A1 and area A2.

In this embodiment, the reading system 10 is not included in the components of the management system 100, but the reading system 10 may be included in the components of the management system 100. Furthermore, in this embodiment, it is assumed that neither the article tag 11 nor the stakeholder tag 21 is included in the components of the management system 100; may be included in the components of That is, the management system 100 may further include the article tag 11, the related party tag 21, and the reading system 10 (reading unit 102).

In this embodiment, the management system 100 includes a learning section 35 as described later. The learning unit 35 is used to perform machine learning on the classifier 310 included in the estimation unit 31 of the management system 100. That is, the learning unit 35 is basically used before the learning of the classifier 310 is completed. Therefore, hereinafter, the management system 100 before the learning of the classifier 310 is completed will be referred to as the "learning system 100." That is, the learning system 100 includes the acquisition section 5, the estimation section 31, the input reception section 34, and the learning section 35. The estimating unit 31 receives as input the read information obtained by acquiring the article information by the acquiring unit 5, and uses the classifier 310 to estimate transportation information regarding the transportation of the article 1 in the predetermined area A1. The learning unit 35 causes the classifier 310 to perform machine learning using the information input at the input receiving unit 34 as training data.

(2.1) Reading System First, the reading system 10 will be explained using FIGS. 1 and 2. The reading system 10 includes an antenna 101, a reading section 102, and a communication section 103, as shown in FIG. In this embodiment, the reading system 10 is configured by accommodating an antenna 101, a reading section 102, and a communication section 103 in one housing. The reading system 10 is installed in area A1 by installing this housing on a wall in area A1.

The antenna 101 receives radio waves (radio waves used as a medium for wireless communication between the reading unit 102, the article tag 11, and the person tag 21). In this embodiment, the antenna 101 is used not only for receiving radio waves but also for transmitting radio waves. The antenna 101 may be a circularly polarized wave antenna or a linearly polarized wave antenna, but in this embodiment, as an example, it is configured with a circularly polarized wave antenna. The antenna 101 is, for example, a microstrip antenna (patch antenna) in which a conductive path is formed on a printed wiring board.

The reading unit 102 performs a reading process of reading article information and stakeholder information by performing wireless communication using radio waves with the article tag 11 attached to the article 1 and the stakeholder tag 21 possessed by the stakeholder 2. It is a device that executes. The reading unit 102 uses the antenna 101 to perform wireless communication with the article tag 11 and the person tag 21 .

In this embodiment, the reading unit 102 is a reader that constitutes an RFID (Radio Frequency Identification) system. That is, the reading unit 102 performs wireless communication with the article tag 11 and the person tag 21 in the RFID system, thereby reading the article information stored in the article tag 11 and the information stored in the person tag 21. Obtain information on related parties.

Each of the article tag 11 and the stakeholder tag 21 is, for example, a passive RF (Radio Frequency) tag. The article tag 11 has a memory that stores at least article information. Further, the related person tag 21 has a memory that stores at least related person information. Here, a plurality of article tags 11 are associated with a plurality of articles 1 on a one-to-one basis. The article tag 11 stores article information about the corresponding article 1, and the article tag 11 is attached to the corresponding article 1. Furthermore, a plurality of stakeholder tags 21 are associated with a plurality of stakeholder tags 2 on a one-to-one basis. The stakeholder tag 21 stores stakeholder information about the corresponding stakeholder 2, and the stakeholder tag 21 is possessed by the corresponding stakeholder 2.

The article tag 11 only needs to be attached to the article 1 in a state that it can be handled integrally with the article 1, and there are various ways in which the article tag 11 can be attached to the article 1. In this embodiment, as an example, the article tag 11 is in the form of a sticker and is attached to the article 1. In addition, the article tag 11 may be connected to the article 1 with a string or the like, may be integrated into the packaging material of the article 1, may be embedded in the article 1, or may be embedded in the article 1. may be incorporated into. Furthermore, the article tag 11 may be formed directly by printing on the surface of the article 1 itself or the packaging material of the article 1, for example, by using a technique such as a coated semiconductor.

The person concerned tag 21 only needs to be possessed by the person concerned 2, and there are various specific ways in which the person tag 21 is possessed by the person concerned 2. In this embodiment, as an example, the person concerned tag 21 is in the form of a sticker and is attached to the clothes of the person concerned 2. In addition, the person tag 21 may be connected to the person's belongings (for example, a mobile terminal or a bag) with a string or the like, or may be embedded in the person's belongings, or may be embedded in the person's belongings. May be incorporated.

The reading unit 102 transmits radio waves from the antenna 101, and performs wireless communication using radio waves as a communication medium with the article tag 11 and stakeholder tag 21 activated by the radio waves. In this embodiment, the reading unit 102 receives information including at least article information from the article tag 11 when wireless communication with the article tag 11 is established. Further, the reading unit 102 receives information including at least related person information from the related person tag 21 when wireless communication with the related person tag 21 is established.

The communication unit 103 includes, for example, a wired communication module. The communication unit 103 communicates with the communication unit 5 of the management system 100 via the communication path N1 by a wired communication module using a signal line such as I2C.

(2.2) Management System Next, the management system 100 of this embodiment will be described using FIGS. 1 and 2. In this embodiment, the management system 100 is configured by accommodating a main computer 3, a communication section (obtaining section) 5, a display section 51, and an input section 52 in one housing. The management system 100 and the reading system 10 are installed in the area A1 by installing this casing on a wall in the area A1. As shown in FIG. 1, the management system 100 includes a main computer 3, a communication section 5, a display section 51, and an input section 52. In this embodiment, the display unit 51 and the input unit 52 are both included in the components of the management system 100, but neither of them need be included in the components of the management system 100.

Furthermore, a first storage section 41 , a second storage section 42 , a third storage section 43 , and a fourth storage section 44 are connected to the management system 100 . In this embodiment, none of the first storage unit 41 to fourth storage unit 44 are included in the components of the management system 100, but any of them may be included in the components of the management system 100. .

In this embodiment, the display section 51 is configured with a touch panel display. Therefore, in this embodiment, the display section 51 also functions as the input section 52 that receives operation input from the person concerned 2. Of course, the display section 51 may be a simple display. That is, the display section 51 may be separate from the input section 52.

The communication unit (acquisition unit) 5 includes, for example, a wired communication module and a wireless communication module. The communication unit 5 communicates with the communication unit 103 of the reading system 10 via the communication path N1 using a wired communication module. Furthermore, the communication unit 5 communicates with the terminal 6 via a network N2 such as the Internet, using a wireless communication module, for example, through optical wireless communication using light such as infrared rays or visible light, or wireless communication using radio waves as a medium. connect. The communication unit 5 may be connected to the network N2 via a communication device such as a router.

As already mentioned, the communication unit (acquisition unit) 5 acquires article information. That is, the communication unit 5 acquires article information by communicating with the communication unit 103 of the reading system 10 via the communication path N1. Further, in this embodiment, the communication unit (acquisition unit) 5 acquires related person information. The stakeholder information is read from the stakeholder tag 21 by the reader 102 performing wireless communication using radio waves with the stakeholder tag 21 carried by the stakeholder 2 involved in transporting the article 1 . That is, the communication unit 5 acquires the related person information by communicating with the communication unit 103 of the reading system 10 via the communication path N1.

The main computer 3 is a computer system mainly composed of one or more processors and memory as hardware. In the main computer 3, various functions are realized by executing programs recorded in the memory using one or more processors. The program may be pre-recorded in the memory of the main computer 3, or may be provided via a telecommunications line, or may be provided by being recorded on a non-transitory recording medium readable by the computer system, such as an optical disk or a hard disk drive. It's okay.

The main computer 3 includes an estimation section 31, a notification section 32, a display control section 33, an input reception section 34, and a learning section 35. Specifically, the main computer 3 executes a predetermined program using one or more processors to perform each function of the estimation section 31, the notification section 32, the display control section 33, the input reception section 34, and the learning section 35. Realize.

The estimation unit 31 estimates transportation information regarding transportation of the article 1 in the area A1 based on the read information obtained by acquiring the article information in the communication section (acquisition section) 5. In this embodiment, the estimation unit 31 includes a machine-learning classifier 310 that receives read information as input and estimates transportation information. The classifier 310 is obtained by machine learning using read information as input data. The classifier 310 may include, for example, a linear classifier such as an SVM (Support Vector Machine), a classifier using a neural network, or a classifier generated by deep learning using a multilayer neural network. .

In this embodiment, the classifier 310 is a classifier using a trained neural network. The trained neural network may include, for example, a CNN (Convolutional Neural Network) or a BNN (Bayesian Neural Network). The estimation unit 31 is realized by mounting a trained neural network on an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array).

The reading information input to the estimating unit 31 (classifier 310) is, for example, the variance of the number of times the article tag 11 is read by the reading unit 102, the intensity (RSSI value) of the received radio wave, or the variance of the intensity, or the received This may include the phase of the radio wave or the dispersion of the phase. When the read information includes a variance in the number of readings, a variance in the intensity of radio waves, or a variance in the phase of the radio waves, it can be said that the estimation unit 31 estimates the conveyance information using the change in the read information.

In this embodiment, the estimation unit 31 (classifier 310) estimates transportation information based on the read information and related person information. More specifically, in this embodiment, the estimation unit 31 (classifier 310) estimates transportation information using the correlation between read information and related person information. The related party information includes, for example, the dispersion of the number of times the related party tag 21 is read by the reading unit 102, the intensity (RSSI value) of the received radio waves or the dispersion of the intensity, or the phase of the received radio waves or the dispersion of the phase. obtain. The "correlation between read information and stakeholder information" is, for example, the covariance between the number of times the article tag 11 is read by the reader 102 and the number of times the stakeholder tag 21 is read by the reader 102. or may include a correlation coefficient. In the present embodiment, two data (explanatory variables), the intensity of radio waves (RSSI value) received from the article tag 11 and the correlation coefficient, are input to the estimation unit 31 as read information and related party information. Ru.

Here, as an example, we will use FIG. I will explain. In FIG. 3, the vertical axis represents the correlation coefficient, and the horizontal axis represents the RSSI value. Moreover, a plurality of dots plotted in FIG. 3 each represent data based on read information acquired by the communication unit (acquisition unit) 5 during a predetermined period. Furthermore, line B1 in FIG. 3 represents a separation plane that classifies transportation information (estimated information) into two patterns. Line B1 is determined in advance by subjecting the classifier 310 to machine learning.

In the example shown in FIG. 3, the estimating unit 31 (classifier 310) determines that if the dot is located in an area above line B1, that is, if the correlation coefficient and RSSI value are relatively large, article 1 is separated from area A1. It is presumed that it was taken out. On the other hand, if the dot is located in a region below the line B1, that is, if the correlation coefficient and RSSI value are relatively small, the estimation unit 31 (classifier 310) determines whether the article 1 is to be carried in or carried out. I presume that there is no. The latter situation may occur, for example, when the article 1 is present in the vicinity of the reading unit 102, but the person concerned 2 is moving without carrying the article 1.

When the estimation unit 31 estimates that one or more articles 1 are being transported, the notification unit 32 notifies a predetermined notification destination. In this embodiment, the notification unit 32 notifies the management system 100 itself and the terminal 6 of the person concerned 2 who has the person tag 21 as the notification destination. In other words, the notification unit 32 notifies at least the article information to the notification destination directly or indirectly associated with the person concerned tag 21 based on the person information acquired by the acquisition unit 5 .

The notification unit 32 transmits, to the management system 100 itself, a signal including a command to display related person information on the display unit 51 of the management system 100. That is, the notification unit 32 transmits to the notification destination a signal that causes the display unit 51 of the notification destination (management system 100) to display at least related person information. As a result, a list of names including the person concerned 2 is displayed on the display unit 51, as shown in FIG. 4, for example (see the area indicated by the dashed line 510 in the figure).

This list displays the names of the holders of one or more related person tags 21 (that is, the related person 2) who can be the notification destination. That is, the communication unit (acquisition unit) 5 obtains not only the party information of the party 2 who is actually transporting the article 1, but also information on other party tags 21 that are moving in the same way as the party 2. In some cases, information related to the holder may be obtained. In this case, the display unit 51 will display related party information of a plurality of related parties 2 as described above. Therefore, the person concerned 2 who is actually transporting the article 1 performs an operation to select his or her own name from among the names of the plurality of persons concerned 2 displayed on the display section 51 (here, the person concerned 2 who is actually transporting the article 1 (touching the touch panel display, which functions as a The related person information of the selected related person 2 is transmitted to the main computer 3 via the communication path N1. Thereby, the main computer 3 is able to link and manage the article information of the article 1 and the person information of the person concerned 2 who is actually transporting the article 1. Note that if there are a plurality of persons involved 2 who are actually transporting the article 1, an operation may be performed to select the names of the plurality of persons concerned 2 using the input section 52.

Then, the notification unit 32 transmits an email to the terminal 6 owned by the person concerned 2 selected by the input unit 52 via the network N2. The email contains a dedicated message for confirming the article information of one or more articles 1 acquired by the communication section (acquisition section) 5 and the transportation information of one or more articles 1 estimated by the estimation section 31. Contains the URL (Uniform Resource Locator) of the website. By accessing the URL included in the email, the person concerned 2 who received the email can check the article information and transportation information regarding one or more articles 1 that are actually being transported by looking at the display unit 60. can. The content displayed on the display section 60 will be explained in detail in the explanation of the display control section 33, which will be described later.

The display control unit 33 causes the display unit 60 to display a combination of the article information acquired by the communication unit (acquisition unit) 5 and the transportation information estimated by the estimation unit 31. In this embodiment, the display unit 60 is a display of a terminal 6 (here, a smartphone) owned by the person concerned 2. The display control unit 33 generates information on the combination of article information and transportation information to be displayed on the display unit 60 when the person concerned 2 accesses the above-mentioned dedicated website. Thereby, the above-mentioned combination can be displayed on the display section 60 by the person concerned 2 operating the terminal 6 and accessing the URL of the dedicated website included in the email.

Specifically, as shown in FIG. 5, the display unit 60 includes a first area C1, a second area C2, a third area C3, a fourth area C4, a fifth area C5, and a sixth area C5. Area C6 is displayed. Note that the broken lines in FIG. 5 are only drawn to clarify the range occupied by each of the first area C1 to the sixth area C6, and are not actually displayed on the display unit 60.

In the first area C1, a button C11 having a character string "Decision" is displayed. When the person concerned 2 selects the button C11, the article information and transportation information for each article 1 are determined, and the determined article information and transportation information for each article 1 are transmitted to the main computer 3 via the network N2.

Estimated information for each article 1 is displayed in the second area C2. Here, as estimated information, three radio buttons C21 to C23 corresponding to "not applicable," "bring in," and "take out," respectively, are displayed in the second area C2 for each article 1. "Not applicable" indicates that the article 1 is not subject to either loading or unloading in area A1. "Bringing in" indicates that the article 1 is to be brought into area A1. “Take out” indicates that the article 1 is to be taken out from area A1.

At the time when the person concerned 2 accesses the URL, radio buttons C21 to C23 corresponding to the state of the article 1 estimated by the estimation unit 31 are selected in the second area C2. Then, the person concerned 2 can modify the estimation result by the estimation unit 31 by selecting one of the radio buttons C21 to C23 depending on the actual state of the article 1.

For example, it is assumed that the "take out" radio button C23 is selected for any article 1 when the person concerned 2 accesses the URL. Here, if the person concerned 2 has not taken out the article 1 (that is, has not taken it out from the area A1), the person concerned 2 can check the estimation result by the estimation unit 31 by selecting the "Not applicable" radio button C21. It is possible to correct it. Note that if the estimation result by the estimation unit 31 is correct, the person concerned 2 does not need to operate the radio buttons C21 to C23.

In the third area C3, the name of the article 1 and the product number of the article 1 are displayed for each article 1. The name and product number of the article 1 are included in the article information acquired from the article tag 11. The name of the article 1 may be a simplified name such as "Book 1" shown in FIG. 5, or a more specific name (for example, if the article 1 is a book, the title of the book, etc.) It may be.

In the fourth area C4, the moving location of the article 1, that is, the destination or destination of the article 1 is displayed for each article 1. In the example shown in FIG. 5, a drop-down list C41 including a plurality of preset movement locations is displayed for each article 1 in the fourth area C4. The person concerned 2 can input the actual moving location of the article 1 by selecting the moving location for each article 1 from the drop-down list C41.

Note that the display control unit 33 may display a combo box that is a combination of a drop-down list and a text box in the fourth area C4 instead of the drop-down list C41. With this aspect, the person concerned 2 can not only select the actual moving location of the article 1 from a plurality of predetermined moving locations, but also directly input the moving location.

In the fifth area C5, a text box C51 in which an EPC-URI (Uniform Resource Identifier) printed on the article tag 11 of the article 1 can be input is displayed for each article 1. For example, if there is an article 1 that is not displayed on the display section 60 (that is, an article 1 for which the acquisition section 5 has not acquired article information), the person concerned 2 enters the EPC-URI of this article 1 in the text box C51. Article 1 can be added by inputting the information.

In the sixth area C6, the number of articles 1 estimated by the estimation section 31, in other words, the number of article tags 11 whose article information has been read by the reading section 102 is displayed. In the example shown in FIG. 5, the number of articles 1 is displayed in the sixth area C6, regardless of whether or not they are to be carried in or carried out in area A1. The number of 1 may be displayed.

In the present embodiment, the display control unit 33 displays the article information and estimated information of each of the plurality of articles 1 on the display according to the level of the estimated probability that the articles are to be carried in or carried out in the predetermined area A1. 60 may be displayed. In the following, items to be brought into or taken out of area A1 are "laptop PC," "camera lens," "experimental equipment," "trolley," "book 1," "book 2," and "office chair." Assume that the probability that it is estimated to be true is low. Further, in the following, it is assumed that the estimating unit 31 estimates that the "notebook PC", "camera lens", "experimental equipment", and "trolley" are the articles 1 to be carried out from the area A1. Further, in the following, it is assumed that the estimating unit 31 estimates that "Book 1", "Book 2", and "Office Chair" are items 1 that are neither subject to import nor export in area A1. .

As an example, the display control section 33 may display the article information and estimated information of each of the plurality of articles 1 on the display section 60 in a predetermined positional relationship depending on the level of the estimated probability. Specifically, as shown in FIG. 6, the display unit 60 displays two areas C71 and C72 that are divided into upper and lower regions with a horizontally long line segment C70 as a boundary. In the upper region C71, the article 1 estimated by the estimation unit 31 to be carried out from the area A1 is displayed. In the lower region C72, the article 1 estimated by the estimation unit 31 to be neither to be carried in nor to be carried out in the area A1 is displayed. Furthermore, in both of the two regions C71 and C72, the article 1 with a higher estimated probability is displayed higher. In this aspect, compared to the case where a plurality of articles 1 are randomly displayed on the display section 60 regardless of the above-mentioned probability, the person concerned 2 can grasp whether the estimation result by the estimation section 31 is correct or not. It has the advantage of being easy.

Note that in the display section 60, the two areas C71 and C72 may be divided into left and right sides with the vertically long line segment C70 as a boundary. Moreover, the two areas C71 and C72 may each be divided into two pages. Furthermore, on the display unit 60, the plurality of articles 1 may be displayed in three or more areas depending on the level of estimated probability.

Further, as an example, the display control unit 33 may cause the display unit 60 to display the article information and estimated information of each of the plurality of articles 1 in a visually differentiated manner depending on the level of the estimated probability. Specifically, as shown in FIG. 7, a plurality of articles 1 are displayed on the display unit 60 in a color-coded manner depending on the level of estimated probability. In FIG. 7, the range represented by dots indicates that the color of the range is a color other than black (for example, blue). Articles 1 estimated by the estimation unit 31 to be carried out from area A1 are shown in colors other than black. Moreover, the article 1 estimated by the estimation unit 31 to be neither to be carried in nor to be carried out from the area A1 is shown in black. Furthermore, the higher the estimated probability, the darker the color (in FIG. 7, the higher the dot density, the darker the color). In this aspect, it is easier for the person concerned 2 to grasp whether the estimation result by the estimation unit 31 is correct or not, compared to the case where the plurality of articles 1 are all displayed in the same manner regardless of the above-mentioned high or low probability. , there is an advantage.

The input accepting unit 34 accepts input for evaluating the estimation result of the estimating unit 31. In this embodiment, the input is performed on a terminal 6 (here, a smartphone) owned by the person concerned 2. Specifically, the input is to select the button C11 displayed on the display unit 60, select any of the radio buttons C21 to C23, select the movement location from the drop-down list C41, and select the text This may include inputting a character string into box C51. In this embodiment, the input is performed by the person concerned 2 touching the display unit 60 (touch panel display) of the terminal 6. The information input to the terminal 6 is transmitted to the main computer 3 via the network N2, and is received by the input reception unit 34.

As an example, if the transportation information (estimated information) displayed on the display unit 60 is correct, the person concerned 2 selects the button C11. As a result, information indicating that the estimation result by the estimation section 31 is correct is input to the input reception section 34. Further, as an example, if part or all of the transportation information (estimated information) displayed on the display unit 60 is incorrect, the person concerned 2 may choose one of the radio buttons C21 to C23 that indicates the actual state of the article 1. After selecting the radio button, select button C11. As a result, information obtained by correcting the estimation result by the estimation section 31 is input to the input reception section 34.

The learning unit 35 causes the classifier 310 to perform machine learning using the information input at the input receiving unit 34 as training data. “Machine learning” in this disclosure may include relearning performed on the trained classifier 310. The learning unit 35 is composed of one or more processors. The "processor" here may include, for example, a general-purpose processor such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), as well as a dedicated processor specialized for calculations in a neural network.

In this embodiment, the learning unit 35 performs machine learning of the classifier 310 by performing machine learning of a neural network used in the classifier 310. Machine learning is performed using a training dataset. A “learning data set” in the present disclosure refers to a plurality of learning data, in which one learning data is a combination of input information input to the input layer of a neural network and teacher data corresponding to the input information. It is a gathering. In this embodiment, the input information includes, for example, the intensity of radio waves (RSSI value) received from the article tag 11 as read information and a correlation coefficient as related person information. In the learning data, the teacher data is the actual state of the article 1 whose article information has been acquired by the communication section (acquisition section) 5.

The learning unit 35 inputs input information to the input layer of the neural network and executes calculations for each of the plurality of learning data. Then, the learning unit 35 executes backpropagation (error backpropagation method) processing using the output values of a plurality of neurons in the output layer of the neural network and the teacher data. Here, the plurality of neurons in the output layer each correspond to a plurality of states that the article 1 can take. Here, the "multiple states" include a state in which the article 1 is neither carried in nor carried out in the area A1, a state in which the article 1 is subject to be carried into the area A1, and a state in which the article 1 is carried out from the area A1. and the state that is the target of the process.

In the backpropagation process, the learning unit 35 updates the weighting coefficient of the neural network so as to maximize the output value of the neuron corresponding to the teacher data among the plurality of neurons in the output layer. The learning unit 35 optimizes the weighting coefficients of the neural network by performing backpropagation processing on all the learning data. This completes the learning of the neural network.

Each of the first storage unit 41 to fourth storage unit 44 is configured by at least one of a non-temporary recording medium such as a hard disk, and a non-temporary recording medium such as a rewritable non-volatile semiconductor memory, as an example. .

The first storage unit 41 stores article information of one or more articles 1 and information regarding the location of the corresponding article 1 in a linked manner. In this embodiment, the article information of one or more articles 1 is based on EPC. In the present embodiment, the first storage unit 41 further stores the destination of the article 1 as information regarding the location of the article 1. The destination of the article 1 is stored in advance in the first storage unit 41, for example, by an operator of the management system 100 performing an input operation. In other words, the destination of the article 1 stored in the first storage unit 41 before the article 1 is transported is the destination to which the article 1 is scheduled to be transported in the future.

The second storage unit 42 stores article information of one or more articles 1 and related person information of a person concerned 2 who is qualified to handle the corresponding article 1 in a linked manner. That is, the second storage unit 42 stores article information and related person information in advance in association with each other. For example, an operator of the management system 100 or the like performs an input operation to associate article information with related party information.

The third storage unit 43 stores, for example, the identification information of the reading unit 102 that read the article information of the article tag 11, the read article information, and the time at which the article information was read, based on the specifications of EPCIS (EPC Information Service). , are linked and memorized. That is, the third storage unit 43 stores the history of article information read from the article tag 11. The third storage unit 43 may store the received signal strength of the radio wave or the phase of the radio wave at the time when the article information is read from the article tag 11 in association with the article information.

The fourth storage unit 44 stores, for example, the identification information of the reading unit 102 that read the stakeholder information of the stakeholder tag 21, the read stakeholder information, and the time when the stakeholder information was read, based on EPCIS specifications, for example. Link and memorize. That is, the fourth storage unit 44 stores the history of related person information read from the related person tag 21. The fourth storage unit 44 may store the received signal strength of the radio wave or the phase of the radio wave at the time when the related person information is read from the related person tag 21 in association with the related person information.

(3) Operation An example of the operation of the management system 100 of this embodiment will be described below. In this embodiment, the management system 100 also has a relearning function as described above, but is basically used in the inference phase after the learning of the classifier 310 is completed. The management system 100 is used as the learning system 100 in the learning phase until the learning of the classifier 310 is completed.

(3.1) Learning Phase First, a learning phase in which the classifier 310 undergoes machine learning before using the management system 100 will be described. That is, the operation of the learning system 100 will be explained. Machine learning in the learning phase is performed, for example, at a learning center. The learning center uses the learning unit 35 to perform machine learning of the neural network used in the classifier 310. When performing machine learning, the weighting coefficients of the neural network are initialized.

First, when the article information on the article tag 11 attached to the article 1 is read by the reading unit 102 installed in the area A1, the read article information is transmitted to the main computer 3 via the communication path N1. Ru. Thereby, the communication unit (acquisition unit) 5 of the main computer 3 acquires the article information of the article 1 (S1). The communication section 5 acquires the article information of the article 1 every time the article information of the article 1 is read by the reading section 102 . That is, while the article 1 is within the readable range of the reading section 102, the communication section 5 continues to acquire the article information of the article 1.

Similarly, when the reader 102 installed in area A1 reads the stakeholder information of stakeholder tag 21 owned by stakeholder 2, the read stakeholder information is transmitted to the main computer via communication path N1. 3. Thereby, the communication unit (acquisition unit) 5 of the main computer 3 acquires the related person information of the related person 2 (S2). The communication unit 5 acquires the stakeholder information of the stakeholder 2 every time the stakeholder information of the stakeholder 2 is read by the reading unit 102 . That is, while the person concerned 2 is within the readable range of the reading section 102, the communication section 5 continues to acquire the person information of the person concerned 2.

The classifier 310 of the estimation unit 31 of the main computer 3 estimates transportation information of the article 1 based on the read information obtained from the acquired article information and the acquired stakeholder information (S3). In this embodiment, as already described, the read information is the intensity of radio waves (RSSI value) received from the article tag 11, and the related person information is the correlation coefficient. Then, the notification unit 32 of the main computer 3 sends an email containing the URL of a dedicated website for checking the item information and estimated transportation information of the item 1 to a predetermined notification destination (here, the terminal 6). (that is, notify a predetermined notification destination) (S4).

The person concerned 2 who is in possession of the terminal 6 accesses the URL of the dedicated website included in the received email, and confirms the article information and estimated transportation information of the article 1. Then, if the transportation information displayed on the display unit 60 is correct, the person concerned 2 selects the button C11 (see FIG. 5). Further, if part or all of the transportation information displayed on the display section 60 is incorrect, the person concerned 2 selects the radio button representing the actual condition of the article 1 from among the radio buttons C21 to C23. , select button C11. As a result, information on the evaluation of the estimation result by the estimation section 31 is input to the input reception section 34 (S5). Then, the learning unit 35 of the main computer 3 determines the combination of the input information input to the classifier 310 (read information and related person information) and the information input to the input reception unit 34 (that is, teacher data). The data is acquired as one learning data (S6).

Thereafter, if the number of accumulated learning data has not reached a predetermined number (S7: No), the learning unit 35 of the main computer 3 accumulates the learning data by repeating the above processes S1 to S6. do. Then, when the number of accumulated learning data reaches a predetermined number (S7: Yes), the learning unit 35 of the main computer uses the accumulated learning data to develop a neural network used in the classifier 310. Machine learning is performed (S8). For example, when the correct answer rate of the estimation result of the transportation information of the article 1 by the classifier 310 reaches a predetermined value or more, the machine learning of the classifier 310 is completed.

(3.2) Inference Phase Next, the inference phase using the management system 100 after the machine learning of the classifier 310 is completed will be described. It should be noted that the processes S10 to S14 in the inference phase are the same as the processes S1 to S5 in the learning phase, respectively, so a description thereof will be omitted here.

When information obtained by evaluating the estimation result by the estimation unit 31 is input to the input reception unit 34, the main computer 3 stores information regarding the location of the article 1 stored in the first storage unit 41 based on the input information. is updated (S15). Specifically, if there is no correction to the estimation result by the estimation unit 31, the update is performed based on the estimation result by the estimation unit 31. On the other hand, if there is a correction to the estimation result by the estimator 31, it is updated based on the corrected estimation result.

As described above, in this embodiment, the estimation unit 31 estimates transportation information regarding transportation of the article 1. In the present embodiment, the estimation result of the estimating section 31 can be evaluated by receiving input from the person concerned 2 at the input receiving section 34. Therefore, this embodiment has the advantage that even if there is an error in the estimation result of the transportation information by the estimation unit 31, it can be corrected, and as a result, the location of the article 1 can be easily managed.

In other words, if the input receiving unit 34 does not exist, the location of the article 1 is managed based on the estimation result regardless of whether the estimation result of the transport information by the estimation unit 31 is correct or not, so the location of the article 1 cannot be managed correctly. there is a possibility. On the other hand, in this embodiment, since the input reception unit 34 is present, the person 2 who actually transports the item 1 provides feedback on the estimation result of the transport information by the estimation unit 31, so that the location of the item 1 can be managed correctly. It has the advantage of being easy to do.

Furthermore, in this embodiment, the transportation information of the article 1 is estimated using the machine-learning classifier 310. Therefore, in this embodiment, if the classifier 310 is machine-trained using a sufficient number of learning data, the transportation information can be estimated more easily than when using an estimator that does not have the classifier 310. This has the advantage that accuracy can be easily improved. Furthermore, this embodiment has the advantage that transportation information can be easily estimated not only for existing articles 1 managed by the management system 100 but also for newly added unknown articles 1.

(4) Modifications The embodiment described above is just one of various embodiments of the present disclosure. The embodiments described above can be modified in various ways depending on the design, etc., as long as the objective of the present disclosure can be achieved. Further, functions similar to those of the management system 100 may be realized by a management method, a (computer) program, a non-temporary recording medium on which the program is recorded, or the like. Similarly, functions similar to those of the learning system 100 may be embodied in a learning method, a (computer) program, a non-transitory recording medium on which the program is recorded, or the like.

A management method according to one embodiment includes an acquisition step, an estimation step, and an input reception step. The acquisition step is a step of acquiring article information. The article information is information regarding the article 1 that is read from the article tag 11 by the reading unit 102 performing wireless communication with the article tag 11 attached to the article 1 using radio waves as a medium. The estimation step is a step of estimating transportation information regarding transportation of the article 1 in the predetermined area A1 based on the read information obtained by acquiring the article information in the acquisition step. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step.

A learning method according to one embodiment includes an acquisition step, an estimation step, an input reception step, and a learning step. The acquisition step is a step of acquiring article information. The article information is information regarding the article 1 that is read from the article tag 11 by the reading unit 102 performing wireless communication with the article tag 11 attached to the article 1 using radio waves as a medium. The estimation step is a step in which the classifier 310 estimates transportation information regarding transportation of the article 1 in the predetermined area A1 by inputting the read information obtained by acquiring the article information in the acquisition step. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step. The learning step is a step in which the classifier 310 is caused to perform machine learning using the information input in the input receiving step as training data.

A program according to one aspect causes one or more processors to execute the above management method or the above learning method.

Modifications of the above embodiment will be listed below. The modified examples described below can be applied in combination as appropriate.

The management system 100 (or learning system 100) according to the present disclosure includes a computer system, for example, in the main computer 3 or the like. A computer system mainly consists of a processor and a memory as hardware. The function of the management system 100 (or learning system 100) in the present disclosure is realized by a processor executing a program recorded in the memory of the computer system. The program may be pre-recorded in the memory of the computer system, may be provided through a telecommunications line, or may be recorded on a non-transitory storage medium readable by the computer system, such as a memory card, optical disc, hard disk drive, etc. may be provided. A processor in a computer system is comprised of one or more electronic circuits including semiconductor integrated circuits (ICs) or large scale integrated circuits (LSIs). The integrated circuits such as IC or LSI referred to herein have different names depending on the degree of integration, and include integrated circuits called system LSI, VLSI (Very Large Scale Integration), or ULSI (Ultra Large Scale Integration). Furthermore, an FPGA (Field-Programmable Gate Array), which is programmed after the LSI is manufactured, or a logic device that can reconfigure the connections inside the LSI or reconfigure the circuit sections inside the LSI, may also be used as a processor. I can do it. The plurality of electronic circuits may be integrated into one chip, or may be provided in a distributed manner over a plurality of chips. A plurality of chips may be integrated into one device, or may be distributed and provided in a plurality of devices. The computer system herein includes a microcontroller having one or more processors and one or more memories. Therefore, the microcontroller is also composed of one or more electronic circuits including semiconductor integrated circuits or large-scale integrated circuits.

Moreover, it is not an essential configuration for the management system 100 (or the learning system 100) that a plurality of functions in the management system 100 (or the learning system 100) are integrated into one housing. The components of the management system 100 (or the learning system 100) may be distributed and provided in a plurality of housings. Furthermore, at least some of the functions of the management system 100 (or the learning system 100) may be realized by, for example, a server device, a cloud (cloud computing), or the like.

In the above-described embodiment, the management system 100 (learning system 100) is installed in the predetermined area A1, but it is not limited to this, and may be realized by a server located in a remote location away from the facility including area A1. good. In this case, the communication path N1 used for communication between the management system 100 and the reading system 10 is, for example, a network such as the Internet. In other words, in this case, the communication unit 103 of the reading system 10 and the communication unit 5 of the management system 100 perform optical wireless communication using light such as infrared rays or visible light, or wireless communication using radio waves as a medium. , communicate via communication path N1. The communication unit 103 of the reading system 10 and the communication unit 5 of the management system 100 may be connected to the communication path N1 via a communication device such as a router.

In the embodiment described above, the reading unit 102 is not limited to the wall in the predetermined area A1, but may be installed at the gate G1.

In the above-described embodiment, the estimation unit 31 (classifier 310) may receive one data (explanatory variable) as read information, or may receive three or more data (explanatory variables) as read information. good.

In the above-described embodiment, the estimation unit 31 estimates transportation information using the classifier 310 using a trained neural network, but the present invention is not limited to this. For example, the estimation unit 31 may estimate the transportation information by comparing the variance of the number of times the article tag 11 is read by the reading unit 102 with a threshold value.

In the embodiment described above, the notification unit 32 may notify the notification destination using a means other than email, such as an SNS (Social Networking Service) message, an automatic telephone reporting service, or a smart speaker.

In the embodiment described above, the display mode by the display control unit 33 may be able to be changed as appropriate by, for example, the person concerned 2 operating the terminal 6.

In the embodiment described above, the display control unit 33 causes the display unit 60 to display an item 1 for which the probability of being a target for import or export in area A1 is estimated to be lower than a predetermined value (for example, several percent). You don't have to. In this aspect, if the article 1 that is not displayed on the display unit 60 is actually a target for loading or unloading in area A1, the person concerned 2 adds it by inputting the EPC-URI, etc. Is possible.

In the embodiment described above, the information input at the input reception unit 34 does not need to be used as training data in machine learning. For example, the information input at the input reception unit 34 may be used only to modify the estimation result at the estimation unit 31.

In the embodiment described above, the input reception unit 34 may accept voice input by the person concerned 2 using the terminal 6.

In the embodiment described above, in the management system 100, the classifier 310 may be subjected to machine learning (re-learning) using the learning unit 35 in the inference phase as well as in the learning phase. That is, the classifier 310 may use the information input to the input receiving unit 34 as training data in machine learning (relearning) of the classifier 310. In the case of re-learning, the learning unit 35 may execute, for example, processes S6 to S8 in the learning phase. That is, since the handled articles 1 and the environment differ depending on the area A1 where the management system 100 is used, the classifier 310 that has been machine learned in the learning phase may not be able to obtain sufficient estimation results depending on the area A1. . Therefore, by accumulating learning data while using the management system 100 and retraining the classifier 310 using the accumulated learning data, the optimal It is possible to achieve this goal.

In the embodiment described above, the management system 100 can also be used to manage products (item 1) in the retail industry. For example, the management system 100 can track how a product (item 1) purchased by a customer (participant 2) is distributed from time to time.

(summary)
As described above, the management system (100) according to the first aspect includes an acquisition section (5), an estimation section (31), and an input reception section (34). The acquisition unit (5) acquires article information. The article information regarding the article (1) is read from the article tag (11) by performing wireless communication using radio waves with the article tag (11) attached to the article (1) in the reading section (102). It is information. The estimating unit (31) estimates transport information regarding the transport of the article (1) in the predetermined area (A1) based on the read information obtained by acquiring the article information in the acquiring unit (5). The input accepting unit (34) accepts input for evaluating the estimation result of the estimating unit (31).

According to this aspect, there is an advantage that the location of the article (1) can be easily managed.

In the management system (100) according to the second aspect, in the first aspect, the transportation information is estimated information that estimates whether the article (1) is to be carried in or out of the predetermined area (A1). including.

According to this aspect, since it is estimated whether or not the article (1) is being transported, there is an advantage that the location of the article (1) can be easily managed.

In the management system (100) according to the third aspect, in the first or second aspect, the reading information includes the number of times the article tag (11) is read by the reading unit (102), the time interval, the frequency, and the number of times the article tag (11) is read by the reading unit (102). It includes at least one of the intensity of the radio wave and the phase of the received radio wave.

According to this aspect, there is an advantage that the estimation accuracy of transportation information is easily improved.

In the management system (100) according to the fourth aspect, in any one of the first to third aspects, the estimation unit (31) estimates transportation information using changes in the read information.

According to this aspect, there is an advantage that the estimation accuracy of transportation information is easily improved.

In the management system (100) according to the fifth aspect, in any one of the first to fourth aspects, the estimating unit (31) includes a machine-learned classifier ( 310).

According to this aspect, there is an advantage that transportation information can be easily estimated not only for existing articles (1) managed by the management system (100) but also for newly added unknown articles (1). .

In the management system (100) according to the sixth aspect, in the fifth aspect, the classifier (310) uses the information input to the input reception unit (34) as training data for machine learning of the classifier (310). used as

According to this aspect, there is an advantage that the classifier (310) can be retrained even while the management system (100) is in use, and the estimation accuracy of the classifier (310) can be easily improved.

The management system (100) according to the seventh aspect further includes a display control section (33) in the first to sixth aspects. The display control unit (33) causes the display unit (60) to display a combination of the article information acquired by the acquisition unit (5) and the transportation information estimated by the estimation unit (31).

According to this aspect, there is an advantage that the person concerned (2) who looks at the display section (60) can understand whether the estimation result by the estimation section (31) is correct or not.

In the management system (100) according to the eighth aspect, in the seventh aspect, the transportation information is estimated information estimating whether the article (1) is to be carried in or out of the predetermined area (A1). including. There are a plurality of articles (1). The display control unit (33) displays article information and estimated information for each of the plurality of articles (1) depending on the estimated probability that the article is subject to import or export in the predetermined area (A1). According to this aspect, the person concerned (2) who looks at the display section (60) has the advantage that it is easy for the person concerned (2) to understand whether the estimation result by the estimation section (31) is correct or not. .

In the management system (100) according to the ninth aspect, in the eighth aspect, the display control unit (33) displays the article information and estimated information of each of the plurality of articles (1) according to the level of the estimated probability. Displayed on the display unit (60) in a predetermined positional relationship.

According to this aspect, there is an advantage that the person concerned (2) who looks at the display section (60) can easily understand whether the estimation result by the estimation section (31) is correct or not.

In the management system (100) according to the tenth aspect, in the eighth or ninth aspect, the display control unit (33) displays the article information and information of each of the plurality of articles (1) according to the level of the estimated probability. The estimated information is displayed on the display unit (60) in a visually distinct manner.

According to this aspect, there is an advantage that the person concerned (2) who looks at the display section (60) can easily understand whether the estimation result by the estimation section (31) is correct or not.

In the management system (100) according to the eleventh aspect, in any one of the first to tenth aspects, the acquisition unit (5) further acquires related person information. The stakeholder information is retrieved from the stakeholder tag (21) by the reader (102) through radio communication using radio waves with the stakeholder tag (21) possessed by the stakeholder (2) involved in the transportation of the item (1). 21) is information regarding the person concerned (2). The management system (100) further includes a notification section (32). The notification unit (32) notifies at least the article information to the notification destination directly or indirectly associated with the stakeholder tag (21) based on the stakeholder information acquired by the acquisition unit (5). I do.

According to this aspect, there is an advantage that it is easy to prompt the person concerned (2) to confirm the whereabouts of the article (1).

In the management system (100) according to the twelfth aspect, in any one of the first to eleventh aspects, the acquisition unit (5) further acquires related party information. The stakeholder information is retrieved from the stakeholder tag (21) by the reader (102) through radio communication using radio waves with the stakeholder tag (21) possessed by the stakeholder (2) involved in the transportation of the item (1). 21) is information regarding the person concerned (2). The estimation unit (31) estimates transportation information based on the read information and related person information.

According to this aspect, there is an advantage that the accuracy of estimating transportation information is easily improved compared to the case where only read information is used.

In the management system (100) according to the thirteenth aspect, in the twelfth aspect, the estimation unit (31) estimates transportation information using the correlation between the read information and the related party information.

According to this aspect, there is an advantage that the accuracy of estimating transportation information is easily improved compared to the case where only read information is used.

The management method according to the fourteenth aspect includes an acquisition step, an estimation step, and an input reception step. The acquisition step is a step of acquiring article information. The article information regarding the article (1) is read from the article tag (11) by performing wireless communication using radio waves with the article tag (11) attached to the article (1) in the reading section (102). It is information. The estimation step is a step of estimating transportation information regarding transportation of the article (1) in the predetermined area (A1) based on the read information obtained by acquiring the article information in the acquisition step. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step.

According to this aspect, there is an advantage that the location of the article (1) can be easily managed.

A learning system (100) according to a fifteenth aspect includes an acquisition section (5), an estimation section (31), an input reception section (34), and a learning section (35). The acquisition unit (5) acquires article information. The article information regarding the article (1) is read from the article tag (11) by performing wireless communication using radio waves with the article tag (11) attached to the article (1) in the reading section (102). It is information. The estimation unit (31) inputs the read information obtained by acquiring the article information in the acquisition unit (5), and uses the classifier (310) to determine the transportation related to the transportation of the article (1) in a predetermined area (A1). Estimate information. The input accepting unit (34) accepts input for evaluating the estimation result of the estimating unit (31). The learning unit (35) causes the classifier (310) to perform machine learning using the information input at the input receiving unit (34) as training data.

According to this aspect, there is an advantage that the location of the article (1) can be easily managed.

The learning method according to the sixteenth aspect includes an acquisition step, an estimation step, an input reception step, and a learning step. The acquisition step is a step of acquiring article information. The article information regarding the article (1) is read from the article tag (11) by performing wireless communication using radio waves with the article tag (11) attached to the article (1) in the reading section (102). It is information. The estimation step is a step of estimating transportation information regarding transportation of the article (1) in the predetermined area (A1) using the classifier (310) using the read information obtained by acquiring the article information in the acquisition step as input. be. The input receiving step is a step of receiving an input for evaluating the estimation result of the estimation step. The learning step is a step in which the classifier (310) is subjected to machine learning using the information input in the input receiving step as training data.

According to this aspect, there is an advantage that the location of the article (1) can be easily managed.

The program according to the seventeenth aspect causes one or more processors to execute the management method according to the fourteenth aspect or the learning method according to the sixteenth aspect.

According to this aspect, there is an advantage that the location of the article (1) can be easily managed.

The configurations according to the second to thirteenth aspects are not essential to the management system (100) and can be omitted as appropriate.

1 Article 11 Article tag 2 Person concerned 21 Person concerned tag 31 Estimation section 310 Classifier 32 Notification section 33 Display control section 34 Input reception section 35 Learning section 5 Communication section (acquisition section)
60 Display section 100 Management system (learning system)
102 Reading section A1 Predetermined area

Claims (11)

an acquisition unit that acquires article information regarding the article read from the article tag by performing wireless communication using radio waves as a medium with the article tag attached to the article in the reading section;
an estimating unit that estimates transportation information regarding transportation of the article in a predetermined area based on read information obtained by acquiring the article information in the acquisition unit;
an input reception unit that receives input for evaluating the estimation result of the estimation unit;
a display control unit that causes a display unit to display a combination of the article information acquired by the acquisition unit and the transportation information estimated by the estimation unit ,
The transportation information includes estimated information estimating whether the article is to be carried in or carried out in the predetermined area,
The reading information includes at least one of the number of times the article tag is read by the reading unit, the time interval, the frequency, the intensity of the received radio waves, and the phase of the received radio waves,
The said article is plural,
The display control unit displays the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the article is a target for import or export in the predetermined area. display on the display section ;
management system. The estimation unit estimates the transportation information using the change in the read information.
The management system according to claim 1. The estimation unit includes a machine-learned classifier that receives the read information as input and estimates the transportation information.
A management system according to claim 1 or 2. The classifier uses the information input to the input reception unit as training data in machine learning of the classifier.
The management system according to claim 3. The display control section causes the display section to display the article information and the estimated information of each of the plurality of articles in a visually differentiated manner depending on the level of the estimated probability.
The management system according to any one of claims 1 to 4. The acquisition unit is configured to perform wireless communication using radio waves with a stakeholder tag possessed by a stakeholder involved in the transportation of the article by the reading unit, so that a stakeholder related to the stakeholder is read from the stakeholder tag. Get more information,
Further comprising a notification unit that notifies at least the article information to a notification destination directly or indirectly associated with the party tag, based on the party information acquired by the acquisition unit.
The management system according to any one of claims 1 to 5. The acquisition unit is configured to perform wireless communication using radio waves with a stakeholder tag possessed by a stakeholder involved in the transportation of the article by the reading unit, so that a stakeholder related to the stakeholder is read from the stakeholder tag. Get more information,
The estimating unit estimates that the article has been carried out from the predetermined area when a correlation coefficient regarding the correlation between the read information and the related person information and the intensity of radio waves received from the article tag are large.
The management system according to any one of claims 1 to 6. A management method performed by a computer system, the method comprising:
an acquisition step of acquiring article information regarding the article read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading unit;
an estimation step of estimating transportation information regarding transportation of the article in a predetermined area based on read information obtained by acquiring the article information in the acquisition step;
an input receiving step for receiving input for evaluating the estimation result of the estimation step;
a display control step of causing a display unit to display a combination of the article information acquired in the acquisition step and the transportation information estimated in the estimation step;
The transportation information includes estimated information estimating whether the article is to be carried in or carried out in the predetermined area,
The reading information includes at least one of the number of times the article tag is read by the reading unit, the time interval, the frequency, the intensity of the received radio waves, and the phase of the received radio waves,
The said article is plural,
The display control step includes displaying the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the plurality of articles are objects of import or export in the predetermined area. display on the display section;
Management method. an acquisition unit that acquires article information regarding the article read from the article tag by performing wireless communication using radio waves as a medium with the article tag attached to the article in the reading section;
an estimating unit that estimates transportation information regarding transportation of the article in a predetermined area using a classifier, using as input read information obtained by acquiring the article information in the acquisition unit;
an input reception unit that receives input for evaluating the estimation result of the estimation unit;
a display control unit that causes a display unit to display a combination of the article information acquired by the acquisition unit and the transportation information estimated by the estimation unit;
a learning unit that causes the classifier to perform machine learning using information input at the input receiving unit as training data;
The transportation information includes estimated information estimating whether the article is to be carried in or carried out in the predetermined area,
The reading information includes at least one of the number of times the article tag is read by the reading unit, the time interval, the frequency, the intensity of the received radio waves, and the phase of the received radio waves,
The said article is plural,
The display control unit displays the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the article is a target for import or export in the predetermined area. display on the display section;
learning system. A learning method performed by a computer system, the method comprising:
an acquisition step of acquiring article information regarding the article read from the article tag by performing wireless communication using radio waves with the article tag attached to the article in a reading unit;
an estimation step of estimating transportation information regarding transportation of the article in a predetermined area using a classifier, using as input the read information obtained by acquiring the article information in the acquisition step;
an input receiving step for receiving input for evaluating the estimation result of the estimation step;
a display control step of displaying on a display a combination of the article information acquired in the acquisition step and the transportation information estimated in the estimation step;
a learning step of performing machine learning on the classifier using the information input in the input receiving step as training data;
The transportation information includes estimated information estimating whether the article is to be carried in or carried out in the predetermined area,
The reading information includes at least one of the number of times the article tag is read by the reading unit, the time interval, the frequency, the intensity of the received radio waves, and the phase of the received radio waves,
The said article is plural,
The display control step includes displaying the article information and the estimated information of each of the plurality of articles in a predetermined positional relationship depending on the estimated probability that the plurality of articles are objects of import or export in the predetermined area. display on the display section;
How to learn. one or more processors,
executing the management method according to claim 8 or the learning method according to claim 10;
program.

Images