Data extraction device, data extraction system, data extraction method, and data extraction program

Abstract

To reduce the number of man-hour for extracting data related to a work content which changes in a time sequence.SOLUTION: A data extraction apparatus having a processor and a storage device which cooperate with a memory in order to execute a program, holds management data related to metadata in the storage device, the metadata being added to time-sequential sensor data collected from a sensor installed in a work environment based on a condition of the work environment; and the processor newly forms an entry of the management data related to the metadata which is added to the sensor data based on a new condition in the work environment based on an existing entry of the management data.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a data extraction device, a data extraction system, a data extraction method, and a data extraction program.

Background Art

[0002] With the development of IoT (Internet of Things), services that provide value-added solutions by analyzing data collected from devices such as sensors have emerged.

[0003] As an example of the application destination of IoT, there is a factory. When applying IoT to a factory, sensor data is collected from sensors attached to manufacturing equipment and workers in the factory, and the state and efficiency of the manufacturing process of the factory are visualized and analyzed, thereby detecting abnormalities at the manufacturing site, improving production plans and manufacturing processes, and improving the production technology of workers.

[0004] As a specific example of a solution that applies IoT to a factory, there is a solution that associates sensor data collected from factory equipment with configuration data of a production line and analyzes the state and efficiency of the production line of the factory. Here, the configuration data of the production line refers to the connection order of equipment installed on the production line, the types of equipment used, the products to be produced, and the like. Conventionally, the work of associating sensor data with the configuration data of the production line is manual, and a person browses all the data and makes selections, so an increase in the man-hours for data association has become a problem.

[0005] For example, when detecting an abnormality of factory equipment or workers from sensor data, for the analysis and countermeasures of the abnormality, not only the relevant equipment and processes but also the relationship with the previous and subsequent processes and the information of the workers of the relevant equipment are required. The configuration of the production line changes according to the products to be produced, and the workers also change according to the work schedule. It has become a technical problem to reduce the man-hours for data association following these changes and variations.

[0006] In response to this technical problem, Patent Document 1 discloses an information processing system that utilizes device data recorded in a database according to data definitions that can be updated at any time. The version number of the device's data definition is managed in association with time, and a data definition corresponding to a specified period is provided.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] However, the data definitions described in Patent Document 1 are information on the position and state of the device body, and do not disclose a mechanism for managing information whose relevance changes over time, such as information on surrounding devices and workers.

[0009] Also, in Patent Document 1, the version number is positioned only by time. However, in the solution for analyzing the state and efficiency of the factory production line described above, not only the change in the configuration of the working environment but also the change in the work content causes the analysis perspective to change. Therefore, simply managing the time-series change of the device's data definition cannot perform the necessary data analysis.

[0010] The present invention has been made in view of the above problems, and an object thereof is to reduce the man-hours required to extract data related to work content that changes over time.

Means for Solving the Problems

[0011] A typical example of the invention disclosed in the present invention is as follows. That is, a data extraction device having a processor and a storage device that execute a program in cooperation with a memory, wherein management data regarding metadata to be added to time-series sensor data collected from sensors installed in a working environment based on the conditions of the working environment is held in the storage device, and the processor newly creates an entry of the management data regarding the metadata to be added to the sensor data based on new conditions of the working environment based on the existing entry of the management data.

Effect of the Invention

[0012] According to the present invention, by performing generation management of metadata according to time and work content, it becomes easy for a data analyst to select relevant data from a large amount of various operation data by using time and work content as triggers, and the man-hours required to extract data related to the work content that changes over time can be reduced. In addition, based on the similarity between past metadata and the corresponding work content, new-generation metadata can be automatically created using past metadata, further reducing man-hours. Problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention is not to be construed as being limited to the description of the embodiments shown below. It will be easily understood by those skilled in the art that the specific configuration can be changed without departing from the spirit or gist of the present invention.

[0015] In the following description, the same or similar configurations or functions are denoted by the same reference numerals, and redundant descriptions are omitted. Also, even when different branch numbers are assigned to the same reference numeral, if they are not distinguished, the reference numeral without a branch number is used.

[0016] <Example of the working environment of the embodiment> FIG. 1 is an explanatory diagram showing an example of a working environment. The working environment 100 has two production lines 181-1 and 181-2. In each production line 181, a device 131 is arranged as a manufacturing device. In the production line 181-1, four devices 131-1, 131-2, 131-3, and 131-4 are arranged to manufacture the product 191-1. In the production line 181-2, four devices 131-5, 131-6, 131-7, and 131-8 are arranged to manufacture the product 191-2.

[0017] The device 131 has a built-in sensor for measuring temperature, current, etc., and has a function of outputting the value of the sensor as sensor data together with the identification information and time information of the device.

[0018] In the working environment 100, eight workers 141-1 to 141-8 are arranged for each device 131 to perform operations such as operation and inspection of each device 131.

[0019] The device 131 is connected to a PLC 151 (Programmable Logic Controller) via an in-factory network 171.

[0020] The PLC 151 is a device for controlling and managing the device 131. In this embodiment, it has a role of collecting sensor data of the sensor built in the device 131. The PLC 151 is connected to an IoT gateway 161 via another in-factory network 172.

[0021] The IoT gateway 161 transmits information such as sensor data acquired from the connected PLC 151 to a server constructed on a network 173 outside the factory such as the Internet. The IoT gateway 161 transfers data received from a server constructed on the network 173 outside the factory to the PLC 151. In this embodiment, the IoT gateway 161 has a role of collecting and aggregating the data of the PLC 151 and transmitting it to the metadata extraction device 101.

[0022] <Hardware Configuration of Metadata Extraction Device 101> FIG. 2 is a block diagram showing a hardware configuration example of the metadata extraction device 101. The metadata extraction device 101 is provided, for example, as an enhanced function of the data processing device 102. The metadata extraction device 101 includes a processor 201, a storage device 202, an input device 203, an output device 204, and a communication interface (communication IF) 205. The processor 201, the storage device 202, the input device 203, the output device 204, and the communication IF 205 are connected by a bus 206. The processor 201 controls the metadata extraction device 101. The storage device 202 serves as a working area for the processor 201. Also, the storage device 202 is a non-temporary or temporary recording medium that stores various programs and data. Examples of the storage device 202 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), and a flash memory. The input device 203 inputs data. Examples of the input device 203 include a keyboard, a mouse, a touch panel, a numeric keypad, and a scanner. The output device 204 outputs data. Examples of the output device 204 include a display and a printer. The communication IF 205 connects to a network and transmits and receives data.

[0023] <Configuration Example of Metadata Extraction System> FIG. 3 is a block diagram showing a configuration example of the metadata extraction system 200. The metadata extraction system 200 includes a metadata extraction device 101, a data processing device 102, a device 131, a PLC 151, and an IoT gateway 161. The data processing device is also referred to as a data processing platform.

[0024] The IoT gateway 161 is connected to the data processing device 102 and transmits sensor data received from one or more PLCs 151 to the data processing device 102. The PLC 151 transmits sensor data received from one or more devices 131 to the IoT gateway 161. The device 131 measures the state of the device and transmits it to the PLC 151 as sensor data. The sensor data transmitted by the device 131 includes multiple types such as temperature, humidity, vibration, current, position, etc. In addition, the sensor data output by the device 131 includes attached information such as the device identifier of the device 131 and time information indicating the measurement time of the sensor data.

[0025] The metadata extraction device 101 includes a configuration data input unit 311, a configuration data storage unit 312, a metadata extraction unit 313, and a data display unit 314. The configuration data input unit 311 and the data display unit 314 are, specifically, functions realized by causing the processor 201 to execute a program stored in the storage device 202 shown in FIG. 2, for example. The configuration data storage unit 312 and the metadata extraction unit 313 are, specifically, functions realized by using the data stored in the storage device 202 shown in FIG. 2 and causing the processor 201 to execute a program stored in the storage device 202.

[0026] The configuration data input unit 311 has a function of capturing work environment configuration information 301 indicating the types of devices 131 in the work environment 100, the connection relationships between the devices 131, the correspondence between the devices 131 and the workers 141, etc., and inputting it as configuration data 411 to the configuration data storage unit 312.

[0027] In addition, the configuration data input unit 311 has a configuration input screen 401 which is a screen for the work environment administrator 331 to input the configuration of the work environment, captures the configuration of the work environment input by the work environment administrator 331, and has a function of inputting it as configuration data 411 to the configuration data storage unit 312.

[0028] <Work environment configuration information 301> FIG. 4 is an explanatory diagram showing an example of the work environment configuration information 301. FIG. 4 shows the work environment configuration information 301 of the production line 181-1. The example shown in FIG. 4 is the work environment configuration information 301 described in the data format of JSON (JavaScript (registered trademark) Object Notation). In the work environment configuration information 301, the production model (“production_model”) manufactured in the work environment 100 is used as the vertex, and the information of the production line 181-1 (“line”) is under the production model, and the information of the IoT gateway 161-1 (“iot_gateway”) is under the production line 181-1. A hierarchical structure having information of the PLCs 151-1 and 151-2 (“plc”) is described under the IoT gateway 161-1.

[0029] In the work environment configuration information 301, a hierarchical structure having information of the devices 131-1 and 31-2 (“equipment”) under the PLC 151-1 and information of the device 131-3 (“equipment”) under the PLC 151-2 is described.

[0030] As information of each element of the work environment configuration information 301, an identifier (“id”) is essential. Other information may be added arbitrarily.

[0031] In FIG. 4, it is represented in the JSON data format, but it may be expressed in other data formats. For example, a similar expression is possible in the XML (eXtensible Markup Language) format, or an original data format may be defined.

[0032] <Configuration input screen 401 of the configuration data input unit 311> FIG. 5 is an explanatory diagram showing an example of the configuration input screen 401 of the configuration data input unit 311. The configuration input screen 401 is used when the work environment manager 331 inputs the configuration of the work environment. The configuration input screen 401 has a first input field 501 to a fourteenth input field 514 and a registration button 515. When the registration button 515 is pressed, the information input in the first input field 501 to the fourteenth input field 514 is registered in the work environment configuration information 301.

[0033] In the first input field 501, the identification information of the production line 181 can be input. In the second input field 502, the identification information of the product 191 can be input. In the third input field 503, the identification information of the IoT gateway 161 can be input. In the fourth input field 504, the identification information of the production line 181 can be input.

[0034] In the fifth input field 505, the identification information of the PLC 151 can be input. In the sixth input field 506, the identification information of the IoT gateway 161 can be input. In the seventh input field 507, the identification information of the device 131 can be input. In the eighth input field 508, the identification information of the PLC 151 can be input. In the ninth input field 509, the type of process can be input. In the tenth input field 510, the identification information of the device 131 can be input. In the eleventh input field 511, the identification information of the operator 141 can be input. In the twelfth input field 512, the type of process can be input. In the thirteenth input field 513, the type of process can be input. In the fourteenth input field 514, the type of process can be input.

[0035] On the configuration input screen 401, the elements of the identification information input in the second input field 502, the fourth input field 504, the sixth input field 506, and the twelfth input field 512 on the right side are located higher in the work environment configuration information 301. The relevance between the manufacturing model and the production line 181 is registered in the work environment configuration information 301 by the first input field 501 and the second input field 502. The relevance between the manufacturing model and the production line 181 is registered in the work environment configuration information 301 by the third input field 503 and the fourth input field 504.

[0036] The relevance between the IoT gateway 161 and the PLC 151 is registered in the work environment configuration information 301 by the fifth input field 505 and the sixth input field 506. The relevance between the PLC 151 and the device 131 is registered in the work environment configuration information 301 by the seventh input field 507 and the eighth input field 508. The relevance between the device 131 and the type of process is registered in the work environment configuration information 301 by the ninth input field 509 and the tenth input field 510. The relevance between the process and the worker 141 is registered in the work environment configuration information 301 by the eleventh input field 511 and the twelfth input field 512. The adjacency relationship between processes is registered in the work environment configuration information 301 by the thirteenth input field 513 and the fourteenth input field 514.

[0037] The screen configuration of FIG. 5 is merely an example, and another screen configuration may be used. For example, instead of input fields, a screen configuration where selection is made using a pull-down box may be used, or a screen configuration where a JSON format file as shown in FIG. 4 is read may be used.

[0038] <Configuration data 411> FIG. 6 is an explanatory diagram showing an example of the configuration data 411. The configuration data 411 is represented, for example, as a table having a configuration data creation time 601 and a graph format database ID 602 as fields. The configuration data creation time 601 is a predetermined time when the graph format database ID 602 is saved. The configuration data creation time 601 may be at regular intervals or at the time of configuration change. Thus, the graph format database ID 602, that is, the work environment configuration information 301, is registered periodically or irregularly.

[0039] The graph format database ID 602 is an identifier of a graph format database indicating the work environment configuration information 301. Each time the metadata extraction device 101 creates and updates the configuration data 411 in response to the input of the work environment configuration information 301 and the worker configuration information 302, an entry is added to the table of the configuration data 411. By referring to the graph format database ID 602 of the configuration data 411, the content of the work environment configuration information 301 can be confirmed as shown in FIG. 7 described later.

[0040] <Graphical format database of configuration data 411> FIG. 7 is an explanatory diagram showing an example of a graphical format database of configuration data 411. The configuration data 411 represents the content of the work environment configuration information 301 in graphical format data. In the graphical format data 701, the work environment configuration information 301 is represented by connecting the product 191, the production line 181, the IoT gateway 161, the PLC 151, the device 131, the worker 141, and the relevance of the process with arrows.

[0041] <Example of data collection method and metadata> FIG. 8 is an explanatory diagram showing an example of data collection and processing in the data processing device 102. First, the IoT gateway 161 transmits the first data 801 composed of a packet ID, an input data ID, and binary data.

[0042] Next, the data collection unit 321 of the data processing device 102 receives the first data 801, which is binary data, from the IoT gateway 161. The data collection unit 321 transmits the received first data 801 to the data processing unit 322. The data processing unit 322 converts the first data 801 into the second data 802 according to the conversion logic stored in the conversion logic 451. Specifically, for example, the binary data (the first data 801) is converted into string data (the second data 802) including a timestamp, a device ID, a temperature, and a current. The data processing unit 322 transmits the second data 802 to the metadata addition unit 323. The method by which the data processing unit 322 converts the first data 801 is stored in the conversion logic 451 of the data processing unit 322. The conversion logic 451 will be described later.

[0043] Next, the metadata addition unit 323 adds supplementary data necessary for data analysis to the received second data 802 to generate third data 803. In the present embodiment, this supplementary data is defined as metadata. For example, the third data 803 is data obtained by adding, as metadata, data item name 1 (line ID), data item name 2 (responsible process ID), data item name 3 (previous process ID), data item name 4 (next process ID), and data item name 5 (operator ID) to the string data (second data 802) including a time stamp, device ID, temperature, humidity, and current. The third data 803 is transmitted from the metadata addition unit 323 as processed data 804 and stored in the collected data storage unit 324. The metadata added by the metadata addition unit 323 is stored in the metadata 481 of the metadata management unit 326. Explanation of the metadata will be given later.

[0044] This embodiment discloses metadata extraction for creating metadata used in the data processing device 102 from the configuration data 411 held by the configuration data storage unit 312 of the metadata extraction device 101.

[0045] Returning to FIG. 3, the functions and methods for metadata extraction will be described. The metadata extraction unit 313 has a function for creating metadata used in the data processing device 102 from the configuration data 411 held by the configuration data storage unit 312, and includes a metadata creation unit 421, a metadata extraction rule 422, generation management data 423, and a metadata output unit 424.

[0046] The metadata creation unit 421 creates metadata indicating information necessary for analyzing data collected by the data processing device 102 using the configuration data 411 of the configuration data storage unit 312. The metadata extraction rule 422 stores a metadata extraction rule that is a search condition for the configuration data 411 used by the metadata creation unit 421 to create metadata.

[0047] In addition, the metadata creation unit 421 assigns a version number to the metadata in order to manage the generation of the metadata. Information on the version number of the metadata is stored in the generation management data 423.

[0048] The metadata output unit 424 has a function of outputting, to the data processing device 102, the metadata to be added to the data collected by the data processing device 102. The metadata output by the metadata output unit 424 is created by the metadata creation unit 421.

[0049] <Metadata extraction rule 422> FIG. 9 is an explanatory diagram showing an example of the metadata extraction rule 422. The metadata extraction rule 422 is represented by a table having, as fields, an extraction rule ID 901 and a search condition 902. The extraction rule ID 901 is an identifier indicating the extraction rule. The search condition 902 indicates a search condition for acquiring necessary data from the configuration data 411 in the configuration data storage unit 312 in order to create the metadata.

[0050] The search condition 902 is expressed as a combination of a search key 9021 and a search value 9022. The search key 9021 indicates the data type used to search for the necessary data from the configuration data 411. The search value 9022 indicates the data type of the data to be acquired from the configuration data 411. Multiple search values 9022 may be specified for one search key 9021.

[0051] Multiple search conditions 902 may be specified for one extraction rule ID 901. For example, in the entry with the extraction rule ID 901 being R02, the device ID is used as the search key to acquire the line ID, the responsible process ID, the previous process ID, and the subsequent process ID. Then, using the acquired responsible process ID as the next search key, searching for the operator ID is registered as the search condition.

[0052] In the example of Fig. 9, three types of metadata are defined. The first is a metadata definition that specifies a specific device ID and obtains the ID of a line connected to the specified device ID and the ID of the responsible process. Here, the connection relationship means, for example, a relationship that is connected in one hop in the graph-form database of the configuration data 411. For example, this metadata is used during the normal operation of a working environment such as a factory. The second is a metadata definition that specifies a specific device ID and obtains the line ID, responsible process ID, the process ID of the previous process of the responsible process, and the process ID of the next process that are connected to the specified device ID, and also obtains the operator ID responsible for the responsible process ID using the obtained responsible process ID as a search key. For example, when an abnormality occurs in a working environment such as a factory, this metadata is used to understand the detailed situation of the working environment. The third is a metadata definition that specifies a specific device ID and obtains the line ID connected to the specified device ID. For example, this metadata is used during the test run before introducing a device into a production line.

[0053] Depending on the situation of the working environment and the purpose of data analysis, the required data is different, and only the necessary data is collected in order to improve the efficiency of data analysis time and the amount of data storage resources used. As shown in Fig. 9, multiple metadata are defined and used appropriately.

[0054] <Example of Generation Management Method of Metadata> Fig. 10 is an explanatory diagram showing an example of the generation management data 423. The generation management data 423 is represented, for example, as a table having, as fields, a generation management data ID 1001, a registration method 1002, a configuration data creation time 601, a configuration data condition 1003, an event condition 1004, a version number 1005, and an extraction rule ID 901. The generation management data ID 1001 is an identifier indicating generation management data. The registration method 1002 indicates whether the metadata targeted by the generation management data was created manually by the user or automatically created. The configuration data creation time 601 indicates the creation time of the configuration data used by the generation management data. The configuration data condition 1003 indicates the conditions for limiting the extraction range of the metadata targeted by the generation management data ID 1001. The event condition 1004 indicates information on the event using the generation management data and can be arbitrarily defined by the user of the metadata extraction device 101. An example of the definition method of the event condition 1004 is "tag = value", and hereinafter, it will be described with "tag = value". The version number 1005 indicates the version number of the generation management data. The extraction rule ID 901 indicates the extraction rule ID corresponding to the generation management data.

[0055] In the example of FIG. 10, one configuration data condition 1003 is defined for one generation management data ID 1001, but multiple configuration data conditions 1003 may be defined for one generation management data ID 1001. Similarly, in the example of FIG. 10, one event condition 1004 is defined for one generation management data ID 1001, but multiple event conditions 1004 may be defined for one generation management data ID 1001.

[0056] <Metadata management method> The concept of metadata in this embodiment will be described. FIG. 11 is an explanatory diagram of the metadata management method in this embodiment. In this embodiment, metadata is managed based on two elements: time and the content of the work. In this embodiment, the content of the work is defined as an event. Examples of events include, for example, information indicating which product is being manufactured on the production line. Another example is the operating status of the production line, including the normal operating status, the status where an abnormality has occurred, and the status during trial operation.

[0057] In FIG. 11, metadata during normal operation of the production line for manufacturing product P01 is created at time t1, and version number Ver. 1.0 is assigned. At time t2, using the same configuration data as version number Ver1.0, metadata during abnormal times is created by adding data to the normal-time metadata and assigning version number Ver. 1.1. Furthermore, using the same configuration data as version number Ver1.0, version number Ver1.2 is assigned by partially deleting data from the normal-time metadata, and metadata during trial operation is created by assigning version number Ver. 1.2.

[0058] Next, consider creating metadata for the production line for manufacturing product P02 at times t3 and t4. Since the method of creating metadata is the same as that for manufacturing P01 except for changing the product number, it can be created based on the metadata of the production line for manufacturing product P01. In this embodiment, referring to the creation methods of Ver. 1.0, 1.1, and 1.2 of the metadata of product P01, metadata Ver. 1.0, 1.1, and 1.2 of product P02 are automatically created.

[0059] Subsequently, consider creating metadata when the environment of the production line for manufacturing product P01 changes at times t5 and t6. Since the method of creating metadata is the same as Ver. 1.0, 1.1, and 1.2 and only requires re-searching the configuration data, in this embodiment, referring to the creation methods of Ver. 1.0, 1.1, and 1.2 of the metadata of product P01, metadata Ver. 2.0, 2.1, and 2.2 of product P01 are automatically created.

[0060] Thus, in this embodiment, a method is shown for managing metadata based on two elements of time and events, and when creating new metadata, if past metadata can be used, using the past metadata to automatically create metadata.

[0061] <Metadata Extraction Process> Figure 12 is a flowchart showing an example of the metadata extraction process executed by the metadata extraction unit 313. First, the current time, configuration data conditions, and event conditions are input (step S1201).

[0062] Next, the metadata extraction unit 313 refers to the generation management data 423 and searches for an entry corresponding to the conditions input in step S1201 (step S1202).

[0063] If there is one or more similar entries in the generation management data 423 that include the search key for the configuration data condition or the tag name of the event condition that match the conditions input in step S1201 (step S1203: YES), the process proceeds to step S1204. The metadata extraction unit 313 selects the entry that contains the most values of the configuration data condition and the event condition input in step S1201, obtains the extraction rule ID (step S1204), and registers a new entry in the generation management data 423 by adding the conditions input in step S1201 to the similar entry. When registering, the registration method is set to "automatic registration".

[0064] If there are no similar entries that include the search key for the configuration data condition and the tag name of the event condition that match the search conditions (step S1203: NO), the metadata extraction unit 313 accepts the search conditions and version number for creating metadata input by the user (step S1206). Then, the metadata extraction unit 313 registers a new entry of the conditions input in step S1201 in the metadata extraction rule 422, newly obtains the extraction rule ID (step S1207). Then, the metadata extraction unit 313 automatically creates metadata and registers a new entry in the generation management data 423. When registering, the registration method is set to "manual registration".

[0065] <Automatic registration of metadata> FIG. 13 is an explanatory diagram showing an example of automatic registration of metadata performed in step S1202, step S1203, step S1204, and step S1205 of FIG. 12. FIG. 13 shows an example in which "Product ID = P02" is input as the configuration data condition and "Operating status = normal" is input as the event condition for the input of step S1201 in FIG. 12. In step S1202, when referring to the generation management data 423, there are three entries including the search key "Product ID" of the configuration data condition and the tag name "Operating status" of the event condition, with generation management data IDs = V0001, V0002, and V0003. Therefore, step S1203 becomes YES. Next, in step S1204, it is confirmed whether the search value "P02" of the configuration data condition and the value "normal" of the event condition are included. Since only the entry with generation management data ID = V0001 includes "normal", the extraction rule ID901, which is the metadata ID of the entry of V0001, acquires the record of R01.

[0066] Then, in step S1205, as a new entry, the generation management data ID1001 is set to "V0004", the value of the registration method 1002 is set to "automatic registration", the value of the configuration data creation time 601 is set to the value of the configuration data creation time 601 of the latest entry of the configuration data 411. The configuration data condition 1003 is the "Product ID = P02" input in step S1201 of FIG. 12, the event condition 1004 is the "Operating status = normal" input in step S1201 of FIG. 12, the version number 1005 is "Ver. 1.0" because it is a new registration, and the extraction rule ID901 is registered as "R01", which is the extraction rule ID of the entry of V0001.

[0067] In the example of FIG. 13, in step S1204, there was only one entry that most included the values of the configuration data condition and the event condition input in step S1201. However, if there are multiple entries, one entry shall be selected based on conditions such as the largest number of configuration data conditions and event conditions, the latest configuration data creation time, and the latest version number.

[0068] <Example of metadata creation screen> FIG. 14 is an explanatory diagram showing an example of a metadata creation screen 431. The metadata creation screen 431 is a screen used by a metadata developer 341 and has, for example, a configuration data condition input field 1401, an event condition input field 1402, and a creation button 1403. The configuration data condition input field 1401 inputs a search key 14011 and a search value 14012. The event condition input field inputs a tag 14021 and a tag value 14022.

[0069] When the creation button 1403 is pressed, the metadata extraction process shown in the flowchart of FIG. 12 is performed. When performing the manual registration shown in steps S1206 to S1208 of the flowchart of FIG. 12, a screen for the metadata developer 341 to input a metadata extraction method is displayed.

[0070] <Screen for manually registering a metadata extraction method> FIG. 15 is an explanatory diagram showing an example of a screen for the metadata developer 341 to manually register a metadata extraction method. The manual registration screen 432 has, for example, an input field 1501 for the main condition of the metadata extraction condition, an input field 1502 for the subordinate condition of the metadata extraction condition, an input field 1503 for the version number, and a registration button 1504.

[0071] The input field 1501 for the main condition inputs a search key 15011 and a search value 15012. The search value 15012 can be filled in with multiple items by pressing the add button 15013. In FIG. 15, four search values 15012-1 to 15012-4 are input.

[0072] The input field 1502 for the subordinate condition obtains any of the search values 15012 in the input field 1501 for the main condition as a search key 15021, and inputs the obtained search key 15021 and its search value 15022 as search conditions. The search value 15022 can be filled in with multiple items by pressing the add button 15023. Also, the input field 1502 for the subordinate condition can be input multiple times by pressing the add button 15024.

[0073] In the input field 1503 for the version number, enter the version number 15031 when registering in the generation management data 423. When the registration button 1504 is pressed, the manual registration shown in steps S1206 to S1208 of the flowchart in FIG. 12 is performed. The screen configuration in FIG. 15 is merely an example, and another screen configuration may be used. For example, instead of an input field, a screen configuration where selection is made using a pull-down box may be used, or a screen configuration where a file in JSON format or CSV format is read may be used.

[0074] <Confirmation screen 433> FIG. 16 is an explanatory diagram showing an example of the confirmation screen 433 of the metadata extraction unit 313. The confirmation screen 433 of the metadata extraction unit 313 has, for example, a configuration condition input field 1601, an event condition input field 1602, a version condition input field 1603, and a display button 1604. In the configuration condition input field 1601, enter the search key 16011 and the search value 16012. In the event condition input field 1602, enter the tag 16021 and the tag value 16022. For the version condition input field 1603, press either the radio button 16031 for searching for the latest version or the radio button 16032 for specifying a version number. When specifying a version number, enter the version number in the version number input field 16033.

[0075] When the display button 1604 is pressed, the generation management data 423 is referred to based on the input configuration conditions, event conditions, and version conditions, and the corresponding entry is acquired. Regarding the version conditions, when searching for the latest version, the metadata ID of the entry with the latest version number is acquired from among the entries that match the configuration conditions and event conditions. When specifying a version number, an entry that matches the version number entered in the version number input field 16033 is acquired. The screen configuration in FIG. 16 is merely an example, and another screen configuration may be used. For example, instead of an input field, a screen configuration where selection is made using a pull-down box may be used, or instead of radio buttons, a screen configuration using an input field or a pull-down box may be used.

[0076] <Result display when the display button 1604 of the confirmation screen 433 is pressed> FIG. 17 is an explanatory diagram showing an example of the result display when the display button 1604 of the confirmation screen 433 in FIG. 16 is pressed. Based on the search method (search condition 902) corresponding to the extraction rule ID 901 of the acquired generation management data 423 entries from the metadata 481, metadata is extracted from the configuration data 411 and displayed on the screen 1701 in string data such as JSON format. The display example is only an example, and other display methods may be used. For example, it may be displayed in CSV file format or in table format. By the result display in FIG. 17, it is possible to confirm which metadata is extracted for the specified configuration data conditions, event conditions, and version conditions.

[0077] According to the present embodiment, it is possible to manage the configuration data of the working environment whose relationship changes over time, and to follow not only the change of the working environment but also the change in the analysis perspective (such as adding variations of the values of the perspective axis), and automatically create the generation management data 423 regarding the additional metadata.

Example

[0078] Hereinafter, as Example 1 of the present embodiment, the additional processing of metadata in data collection and processing will be described with reference to FIGS. 18 to 24.

[0079] <Metadata 481 of Example 1> FIG. 18 is an explanatory diagram showing an example of the metadata 481. The metadata is represented by a table having, for example, a metadata ID 1800, a configuration data creation time 1801, a registration time 1802, a version 1803, and a metadata body 1804. The metadata ID 1800 is an identifier of the metadata. The configuration data creation time 1801 indicates the creation time of the configuration data that is the basis of the metadata. The registration time 1802 indicates the time when the metadata information was registered in the metadata 481. The version 1803 indicates the version information of the metadata.

[0080] The metadata body 1804 indicates the content of the metadata and has, for example, a search key 18041 and an output metadata value 18042. The search key 18041 indicates the field name used as the search condition for the metadata and its value. The output metadata value 18042 indicates the content of the metadata to be output.

[0081] In the example of FIG. 18, examples of two types of metadata with metadata IDs M0001 and M0002 are shown. For the metadata with metadata ID = M0001, the search key 18041 is the device ID, and as the output metadata value corresponding to the value of the device ID, values of five items, namely the line ID, the responsible process ID, the previous process ID, the subsequent process ID, and the operator ID, are output. For the metadata with metadata ID = M0002, the search key 18041 is the PLC ID, and as the output metadata value, values of three items, namely the device ID, the line ID, and the product ID, are output.

[0082] <Metadata addition process of Example 1> FIG. 19 is a flowchart showing an example of the metadata addition process. First, the data processing developer 351 shown in FIG. 3 inputs data conversion logic to the data conversion logic setting unit 325 of the data processing device 102 (step S1901).

[0083] Next, the data conversion logic setting unit 325 sends the configuration data condition and the event condition to the metadata output unit 424 of the metadata extraction device 101 and requests metadata output (step S1902).

[0084] Next, the metadata output unit 424 of the metadata extraction device 101 refers to the generation management data 423 with the received configuration data conditions and event conditions, and the current time as search conditions, and obtains the corresponding extraction rule ID 901 and version number 1005 (step S1903). In step S1903, the metadata output unit 424 refers to the generation management data 423, and obtains the extraction rule ID 901 and version number 1005 of the entry where the configuration data creation time 601 is the closest before the current time of the search conditions and the configuration data conditions 1003 and event conditions 1004 match the configuration data conditions and event conditions of the search conditions.

[0085] Next, the metadata output unit 424 refers to the metadata extraction rule 422 with the extraction rule ID 901 obtained in step S1903 as the search condition, and obtains the search key 9021 and search value 9022 (step S1904).

[0086] Subsequently, the metadata output unit 424 takes the current time, configuration data conditions, search key 9021 and search value 9022 obtained from the metadata extraction rule 422 as inputs, obtains metadata from the configuration data 411 with the corresponding creation time, and transmits it to the metadata management unit 326 of the data processing device 102 (step S1905). The creation time of the configuration data 411 corresponding to the current time is the creation time of the closest configuration data 411 before the current time.

[0087] The metadata management unit 326 of the data processing device 102 registers the metadata received from the metadata output unit 424 in the metadata 481, and notifies the metadata ID of the registered metadata to the data conversion logic setting unit 325 (step S1906).

[0088] The data conversion logic setting unit 325 combines the input data ID and search key input by the data processing developer 351 in step S1901 described above, and the metadata ID received from the metadata management unit 326 in step S1906 described above, and registers them in the metadata acquisition table 461 of the metadata addition unit 323 (step S1907).

[0089] The data conversion logic setting unit 325 registers the data processing logic input by the data processing developer 351 in the conversion logic 451 of the data processing unit 322 in step S1901 described above, and starts collecting and processing data (step S1908).

[0090] <Logic setting screen 471> FIG. 20 is an explanatory diagram showing an example of a logic setting screen 471 for data collection and processing used in step S1901 of FIG. 19.

[0091] In the example of FIG. 20, it is composed of four input screens: input data ID setting 2001, data processing setting 2002, metadata setting 2003, and data output setting 2004.

[0092] In the input data ID setting 2001 which is the first input screen, the data ID which is an identifier of the data received by the data processing device 102 is input.

[0093] In the data processing setting 2002 which is the second setting screen, the data processing method and the specific processing content are described. In FIG. 20, as an example of the processing method, a method of converting binary data into a character string (JSON) is shown. As input items, information on the byte sequence indicating which byte group of the binary data becomes one data, the field name indicating the content of the data of that byte sequence, and the format of the data are input.

[0094] In FIG. 20, a method of converting binary data into a character string (JSON) is described, but another processing method may also be used. For example, it may be a processing method of converting binary data into comma-separated data. In that case, as input items, information on the byte example indicating the data separation, and the format of each data are examples.

[0095] In the metadata setting 2003 which is the third setting screen, the search key and search value of the configuration data condition are input as the acquisition conditions of the metadata. Also, as the event conditions, the tag and tag value indicating the event are input. This tag and tag value correspond to the event conditions of the generation management data 423 in FIG. 10.

[0096] In the data output setting 2004 which is the fourth setting screen, the type of the collected data in the collected data storage unit 324 which is the output destination of the third data 803 in FIG. 8 is set. In FIG. 20, as an example of the data output setting, an RDB (Relational DataBase) is shown. In the RDB, the table name for storing data is input.

[0097] In FIG. 20, an example of the RDB is shown, but another data output destination may be specified. For example, the method of saving as a file may be specified. In that case, as the setting items, the directory of the file save destination, the save file name, etc. are examples of the input items.

[0098] <Conversion logic 451> FIG. 21 is an explanatory diagram showing an example of the conversion logic 451. The conversion logic 451 is represented, for example, as a table having an input data ID 2101 and a logic 2102 as fields. The input data ID 2101 is an identifier of the input data. The input data ID 2101 is the information input in the input data ID setting 2001 of the logic setting screen 471 in FIG. 20.

[0099] The logic 2102 is information indicating the method of converting the input data. For example, it is information indicating the method of converting binary data into a character string (JSON). The logic 2102 is represented, for example, by a byte sequence 21021, a field name 21022, and a format type 21023.

[0100] The byte sequence 21021 indicates the information of the byte sequence for showing which byte groups of the binary data become one data. The field name 21022 indicates the content of the data of the byte sequence. The format type 21023 indicates the format type of the data. The information of Logic 2102 is the information input in the data processing setting 2002 of the logic setting screen 471 in FIG. 20.

[0101] <Metadata Acquisition Table 461> FIG. 22 is an explanatory diagram showing an example of the metadata acquisition table 461. The metadata acquisition table 461 is represented by a table having, for example, an input data ID 2101, a search key 2201, and a metadata ID 2203. The input data ID 2101 is an identifier of the input data. The input data ID 2101 is the information input in the input data ID setting 2001 of the logic setting screen 471 in FIG. 20.

[0102] The search key 2201 is information used as a search condition when searching for metadata. The search key 2201 is the information input in the metadata setting 2003 of the logic setting screen 471 in FIG. 20. The metadata ID 2202 is an identifier of the metadata.

[0103] The input data ID 2101 and the metadata ID 2202 have an m:n (m and n are positive integers) correspondence relationship. The metadata acquisition table 461 can identify the input data affected by the update of the metadata.

[0104] <Metadata Acquisition Process of Data Processing Device 102> FIG. 23 is an explanatory diagram showing an example of a process for the data processing device 102 to acquire metadata from the metadata extraction device 101. The data conversion logic setting unit 325 transmits the configuration data condition "Product ID = P01" and the event condition "Operation condition = normal" to the metadata output unit 424 of the metadata extraction device 101 and requests metadata output (step S2301).

[0105] Next, the metadata output unit 424 of the metadata extraction device 101 refers to the generation management data 423 using the received configuration data condition "Product ID = P01" and event condition "Operation condition = normal", and the current time as search conditions (step S2302), and obtains the corresponding extraction rule ID "R01" and version number "Ver. 2.0" (step S2303).

[0106] Next, the metadata output unit 424 refers to the metadata extraction rule 422 using the obtained metadata ID "R01" as a search condition (step S2304), and obtains the search key "Device ID" and search value "Line ID, Responsible process ID" (step S2305).

[0107] Subsequently, the metadata output unit 424 inputs the current time, the configuration data condition "Product ID = P01", the search key "Device ID" obtained from the metadata extraction rule 422, and the search value "Line ID, Responsible process ID" (step S2306), and obtains the metadata body from the configuration data 411 at the corresponding time (step S2307).

[0108] Finally, the metadata output unit 424 transmits the obtained metadata body, the configuration data creation time, and the version number "Ver. 2.0" to the metadata management unit 326 of the data processing device 102 (step S2308).

[0109] <Metadata Search Process and Addition Process> FIG. 24 is an explanatory diagram showing an example of the metadata search process and addition process. The metadata search process and addition process are performed by the metadata addition unit 323.

[0110] First, the metadata addition unit 323 receives the second data 802 and refers to the input data ID included in the second data 802. In FIG. 24, "D0001" is obtained as the value of the input data ID.

[0111] Next, the metadata addition unit 323 refers to the metadata acquisition table 461 managed by the metadata addition unit 323, using the input data ID = D0001 as the search condition. In FIG. 24, as the metadata ID corresponding to the input data ID = D0001, "M0001" is obtained, and as the search key corresponding to the input data ID = D0001, "device ID" is obtained. Then, the metadata addition unit 323 obtains the value of the search key "device ID" obtained as the search key corresponding to the input data ID = D0001 from the received second data 802. In FIG. 24, as the value of the device ID included in the second data 802, "E02" is obtained.

[0112] The metadata addition unit 323 refers to the metadata of the metadata management unit 326 using the obtained metadata ID = M0001 and device ID = E02 as the search conditions (step S2301). Each table in FIG. 24 has some omissions. In FIG. 24, as the metadata corresponding to the metadata ID = M0001 and device ID = E02, five items (line ID = L01, responsible process ID = O12, previous process ID = O01, subsequent process ID = O04, operator ID = H02) are obtained (step S2402).

[0113] The metadata management unit 326 adds the metadata obtained in step S2402 to the second data 802 and outputs it as the third data 803.

[0114] <Effect in Example 1> The effects of the invention in this embodiment will be described. Conventionally, it was necessary to manually describe and set metadata sequentially, resulting in variations in the quality of metadata due to personal work and an increase in man-hours due to manual work. On the other hand, in this embodiment, by managing the generation of metadata in terms of time series and events, only a simple search condition screen input as manual work can automatically output the necessary metadata, reducing the time of manual work. Furthermore, when the past metadata and events are similar, new-generation metadata can be automatically created based on the past metadata, realizing further reduction of man-hours.

[0115] In this embodiment, metadata is acquired at the current time, but metadata may also be acquired by specifying a past time period. For example, when inputting data for a certain past period into the data processing device 102, metadata corresponding to the certain period is acquired, and metadata to be added by the metadata addition unit 323 is determined from the set of input data ID, time stamp, and metadata ID. This method is effective when processing data for a certain past period as batch processing.

Embodiment

[0116] Hereinafter, as Example 2, the metadata update process when updating the configuration data 411 in real time will be described.

[0117] <Update of Configuration Data 411 in Example 2> FIG. 25 is an explanatory diagram showing an example of updating the configuration data 411. The configuration data input unit 311 transmits the configuration change time and the configuration change content of the work environment 100 to the configuration data storage unit 312 (step S2501).

[0118] The configuration data storage unit 312 updates the configuration data 411. In the example of FIG. 25, device E04 is replaced with device E09, and process O14 is replaced with process O31.

[0119] <Metadata Update Process> FIG. 26 is a flowchart showing an example of the metadata update process. First, the configuration data input unit 311 acquires a configuration change notification including the configuration change time and the configuration change content of the work environment 100 from the management system of the work environment 100 or the like (step S2601).

[0120] Next, the configuration data input unit 311 transfers the configuration change notification to the configuration data storage unit 312 (step S2602). Next, the configuration data storage unit 312 updates the configuration data 411 based on the configuration change notification received from the configuration data input unit 311 and extracts the updated items (step S2603). Then, the configuration data storage unit 312 transmits the configuration change time and the extracted updated items to the metadata extraction unit 313 (step S2604).

[0121] Subsequently, the metadata extraction unit 313 refers to the generation management data 423 (step S2605), checks the search conditions for all entries, and confirms whether there is an entry related to the update item in the search conditions (steps S2606 to S2608). If there is an entry related to the update item in the search conditions (step S2608: Yes), the update item is copied and newly registered in the generation management data 423 (step S2609).

[0122] Next, the metadata extraction unit 313 acquires the newly registered metadata from the configuration data 411 and transmits it to the metadata management unit 326 (step S2610). Finally, the metadata management unit 326 of the data processing device 102 registers the metadata received from the metadata extraction unit 313 in the metadata 481 (step S2611).

[0123] <Example of updating generation management data 423 accompanying the update of configuration data 411> FIG. 27 is an explanatory diagram showing an example of an update example of the generation management data 423 accompanying the update of the configuration data 411 in FIG. 25. FIG. 27 corresponds to the processing of steps S2606 to S2608 in the flowchart of FIG. 26. In FIG. 27, since there are changes in the devices and processes related to the product P01, three entries of generation management data IDs V0001 to V0003 whose configuration data condition is "device ID = product P01" are to be updated.

[0124] The metadata extraction unit 313 copies three entries with generation management data IDs V0001 to V0003, sets the generation management data ID1001 to V0005, V0006, and V0007 respectively, adds 1 to the version number 1005 to make them Ver. 2.0, Ver. 2.1, and Ver. 2.2 respectively, sets the configuration data creation time 601 to the configuration change time in step S2501 of FIG. 25, and sets the registration method 1002 to "automatic registration", and newly registers them in the generation management data 423.

[0125] <Packet ID table 441> FIG. 28 is an explanatory diagram showing an example of the packet ID table 441. The packet ID table 441 has, for example, a packet ID 2801, a timestamp 2802, and an input data ID 2101.

[0126] The packet ID 2801 indicates an identifier of a packet of input data. The packet ID 2801 is an identifier assigned to each of the data transmitted by the IoT gateway 161. The timestamp 2802 records the time when the data collection unit 321 receives the packet.

[0127] <Metadata switching omission detection process> FIG. 29 is a flowchart showing an example of a process for detecting data of metadata switching omission executed by the data processing device 102.

[0128] First, the metadata management unit 326 transmits the configuration data creation time 1801 and the registration time 1802 of the newly registered metadata by metadata update to the data conversion logic setting unit 325 (step S2901).

[0129] Next, the data conversion logic setting unit 325 refers to the packet ID table 441 and acquires the packet ID 2801 included in "switching period = registration time - configuration data creation time" (step S2902).

[0130] Finally, the data conversion logic setting unit 325 generates a switching omission packet ID list describing the packet ID 2801 acquired in step S2902 (step S2903).

[0131] <Method for generating switching omission packet list> FIG. 30 is an explanatory diagram showing the concept of acquiring the packet ID 2801 in step S2902 of FIG. 29. In FIG. 30, by the metadata update process of FIG. 26, the configuration data creation time 1801 of the newly registered metadata in the metadata 481 is set to "2021 / 4 / 1 13:00:00.600", and the registration time 1802 is set to "2021 / 4 / 1 13:00:01.300".

[0132] In the example of FIG. 30, after "2021 / 4 / 1 13:00:00.600" and before "2021 / 4 / 1 13:00:01.300", three packets of the first data 801-3, 801-4, and 801-5 of the time stamp will leak from the switching. Therefore, a switching leak packet ID list including the packet IDs P0003, P0004, and P0005 of the first data 801-3, 801-4, and 801-5 is generated.

[0133] As an example of how to use the switching leak packet ID list, a process in which the data processing device 102 updates the collected data stored with the information of the old metadata using the switching leak packet ID list will be described.

[0134] <Collection data table 491> FIG. 31 is an explanatory diagram showing an example of a collection data table 491 used by the collection data storage unit 324 to store the collection data.

[0135] The collection data table 491 has, for example, a packet ID 2801, a time stamp 2802, an input data ID 2101, a device ID 3101, a temperature 3102, a current 3103, a line ID 3104, an in-charge process ID 3105, a previous process ID 3106, a subsequent process ID 3107, and an operator ID 3108.

[0136] The packet ID 2801 indicates an identifier of the packet of the input data. The packet ID 2801 is an identifier assigned to each of the data transmitted by the IoT gateway 161. The time stamp 2802 records the time when the data collection unit 321 receives the packet.

[0137] The device ID 3101, the temperature 3102, the current 3103, the line ID 3104, the in-charge process ID 3105, the previous process ID 3106, the subsequent process ID 3107, and the operator ID 3108 are a group of data stored in the third data 803.

[0138] The data processing device 102 extracts the data of the packet IDs described in the switching omission packet ID list from the collected data table 491 and performs rewriting to new metadata.

[0139] <Switching omission data update process> Figure 32 is a flowchart showing an example of a process for updating collected data that has leaked from the switching of metadata executed by the data processing device 102.

[0140] First, the data collection unit 321 transmits the switching omission packet ID list generated by the procedure shown in Figure 30 to the collected data storage unit 324 (step S3201).

[0141] Next, the collected data storage unit 324 acquires the data of the target packet IDs indicated by the switching omission packet ID list received from the data collection unit 321 and transmits it to the metadata addition unit 323 (step S3202).

[0142] The metadata addition unit 323 acquires the input data ID from the data received from the collected data storage unit 324, refers to the metadata acquisition table 461 using the input data ID as a search condition, acquires the corresponding metadata search condition, and transmits it to the metadata management unit 326 (step S3203).

[0143] The metadata management unit 326 acquires the metadata corresponding to the metadata search condition received from the metadata addition unit 323 from the metadata 481 and transmits it to the metadata addition unit 323 (step S3204).

[0144] Then, the metadata addition unit 323 changes the metadata of the data of the target packet ID received from the metadata management unit 326 to the metadata acquired from the metadata management unit 326 and transmits it to the collected data storage unit 324 (step S3205).

[0145] The collected data storage unit 324 updates the data in the collected data table 491 with the data received from the metadata addition unit 323 (step S3206).

[0146] Through the above processing, the data processing device 102 can automatically update the data that has leaked from the metadata switching.

[0147] <Update of collected data leaked from metadata switching> FIG. 33 is an explanatory diagram showing an example of updating collected data that has leaked from metadata switching in the data processing device 102.

[0148] First, the data collection unit 321 transmits the switching leakage packet ID list 3301 to the collected data storage unit 324 (step S3201). In the example of FIG. 33, as the switching leakage packet ID list 3301, in the example shown in FIG. 30, it is assumed to include the packet IDs "P0003, P0004, P0005" of the packets that are the targets of metadata switching leakage.

[0149] Next, the collected data storage unit 324 acquires the data corresponding to the packet IDs "P0003, P0004, P0005" described in the switching leakage packet ID list 3301 from the collected data table 491, and transmits it to the metadata addition unit 323 as the update target packet 3302 (step S3202). The data format of the update target packet 3302 is, for example, the same data format as that of the third data 803.

[0150] Subsequently, the metadata addition unit 323 refers to the content of the data of the received update target packet 3302, transmits the metadata search condition 1901 of the update target packet 3302 to the metadata management unit 326 (step S3203), and acquires the metadata 1902 of the update target packet 3302 from the metadata management unit 326 (step S3204). This processing is the same as the metadata search processing and addition processing in FIG. 24.

[0151] Next, the metadata addition unit 323 updates the content of the received update target packet 3302 with the metadata obtained from the metadata management unit 326, and creates an updated packet 3303. Then, the metadata addition unit 323 transmits the created updated packet 3303 to the collected data accumulation unit 324 (step S3205). The data format of the updated packet 3303 is, for example, the same data format as that of the third data 803.

[0152] In the example of FIG. 33, based on the update example of the configuration data shown in FIG. 25, an example of updating the metadata of the packet with the packet ID P0003 is shown, and the value of the post-process ID 3107 is updated from "O14" to "O31".

[0153] The collected data accumulation unit 324 updates the data in the collected data table 491 with the data of the received updated packet 3303 (step S3206).

[0154] <Example of update of the collected data table 491> FIG. 34 is an explanatory diagram showing an example of the update of the collected data table 491 in the example of FIG. 33. FIG. 34 shows an example of updating the data with the packet IDs P0003, P0004, and P0005 shown in FIG. 30 based on the update example of the configuration data shown in FIG. 25. In this example, the post-process ID 3107 of the data of P0003 and the pre-process ID 3106 of the data of P0004 are updated from "O14" to "O31". There is no change in the data of P0005.

[0155] As described above, it takes time to reflect the update of the metadata, and even when data with switching omissions occurs, by using the information in the packet ID table 441 of the data collection unit 321, it is possible to retrospectively execute the metadata update of the switching omission data automatically later.

[0156] <Effect in the second embodiment> In the future, it is expected that the working environment and layout will be changed seamlessly, and the configuration and content of the working environment will be changed in real time. At that time, it is necessary to change data collection and processing in real time, but it is conceivable that data that cannot keep up with the changes will occur. This process is effective in such cases.

[0157] As an effect of the invention in this embodiment, conventionally, there has been data for which it was impossible to determine whether it was associated with new or old metadata before and after a change in the configuration of the working environment or the content of the work, and it was necessary for an operator with on-site knowledge to check the metadata. However, the present invention can eliminate such work.

[0158] In addition, each time the configuration of the working environment or the content of the work is changed, it is not necessary to update all the metadata every time, and it is also possible to reduce the computational load on the data processing device 102.

[0159] Note that the present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the configurations described. Also, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Further, the configuration of another embodiment may be added to the configuration of one embodiment. Also, for a part of the configuration of each embodiment, other configurations may be added, deleted, or replaced. Also, various processing functions may be appropriately integrated and distributed for processing efficiency and implementation efficiency. Similarly, storage areas for storing various data may be appropriately integrated and distributed for processing efficiency and implementation efficiency.

[0160] As described above, each of the above-described configurations, functions, processing units, processing means, etc. may be realized in hardware by designing a part or all of them, for example, by using an integrated circuit, or may be realized in software by a processor interpreting and executing a program for realizing each function.

[0161] Information such as programs, tables, and files that implement each function can be stored in a storage device such as a memory, a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC (Integrated Circuit) card, an SD card, or a DVD (Digital Versatile Disc). Alternatively, information such as programs, tables, and files that implement each function may be acquired from an external computer equipped with a non-temporary storage device through communication via the communication IF205. Alternatively, information such as programs, tables, and files that implement each function may be recorded on a non-temporary recording medium and acquired by being read by a medium reading device.

[0162] In addition, control lines and information lines show those considered necessary for explanation and do not necessarily show all control lines and information lines required for implementation. In practice, it may be considered that almost all components are interconnected.

Explanation of Signs

[0163] 100: Working environment, 101: Metadata extraction device, 102: Data processing device, 131: Equipment, 141: Operator, 161: IoT gateway, 181: Production line, 200: Metadata extraction system, 301: Working environment configuration information, 311: Configuration data input section, 312: Configuration data storage section, 313: Metadata extraction section, 314: Data display section, 321: Data collection section, 322: Data processing section, 323: Metadata addition section, 324: Collected data storage section, 325: Data conversion logic setting section, 326: Metadata management section, 401: Configuration input screen, 411: Configuration data, 421: Metadata creation section, 422: Metadata extraction rule, 423: Generation management data, 424: Metadata output section, 431: Creation screen, 432: Manual registration screen, 433: Confirmation screen, 441: Packet ID table, 451: Conversion logic, 461: Metadata acquisition table, 471: Logic setting screen, 481: Metadata, 491: Collected data table

Claims

1. A data extraction device having a processor and a storage device that execute a program in cooperation with a memory, managing, based on a version number assigned to metadata added to time-series sensor data collected from sensors installed in a work environment according to conditions of the work environment, generation management data including configuration data conditions indicating a configuration of the work environment and event conditions indicating work content in the work environment, and holding the generation management data in the storage device, wherein the processor receives an input of new conditions of the work environment including a current time, the configuration data conditions, and the event conditions, refers to the generation management data, and determines whether there is an existing generation entry of the generation management data including the configuration data conditions or the event conditions that match the configuration data conditions or the event conditions included in the new conditions, and when there is an existing generation entry, adds the new conditions of the work environment to the existing generation entry and newly creates an entry of the generation management data related to the metadata added to the sensor data A data extraction device characterized by the above.

2. The data extraction device according to claim 1, holding configuration data indicating a configuration of the work environment in the storage device, wherein the generation management data includes an identifier indicating a method for extracting metadata, and the processor acquires the identifier from the generation management data based on a current time and conditions of the work environment that is a source of the sensor data, and extracts metadata to be added to the sensor data from the configuration data based on the method for extracting the metadata corresponding to the acquired identifier A data extraction device characterized by the above.

3. The data extraction device according to claim 2, wherein the processor acquires the identifier from an entry of the generation management data including the specified event conditions A data extraction device characterized by the above.

4. The data extraction device according to claim 3, wherein the processor acquires the identifier from an entry of the generation management data including a specified time and the configuration data conditions A data extraction device characterized by the above.

5. The data extraction device according to claim 1, wherein the processor Create a new generation entry of the generation management data for the metadata to be added to the sensor data in response to an update of the configuration data condition or the event condition, based on the entry of the existing generation. A data extraction device characterized by the above. **Claim 6** The data extraction device according to claim 4, The processor Displays on a display device a screen for receiving the specification of the configuration data condition, the event condition, and the version number, Obtains the identifier from the entry of the generation management data corresponding to the received specification, extracts metadata from the configuration data based on the extraction method of the metadata corresponding to the obtained identifier, and displays it on the screen in a predetermined format A data extraction device characterized by the above. **Claim 7** A data extraction device according to any one of claims 1 to 6, and A data processing device that adds the metadata to the sensor data based on the generation management data created by the data extraction device, having The processor Determines whether there is an update of the metadata to be added to the sensor data in response to an update of the configuration of the working environment, and when there is an update of the metadata, notifies the data processing device of the update time of the metadata and the updated metadata A data extraction system characterized by the above. **Claim 8** The data extraction system according to claim 7, The data processing device Manages the correspondence between the sensor data and the metadata added to the sensor data, Identifies the sensor data corresponding to the updated metadata notified from the data extraction device based on the correspondence A data extraction system characterized by the above. **Claim 9** The data extraction system according to claim 7, The data processing device Identifies the sensor data collected from the sensor during the period from the time when the configuration of the working environment is updated to the time when metadata is added to the sensor data based on the updated configuration of the working environment, Reassigns metadata to the identified sensor data based on the updated configuration of the working environment A data extraction system characterized by the above. **Claim 10** A data extraction method executed by a data extraction device, wherein The data extraction device Manages, based on a version number given to metadata added to time-series sensor data collected from sensors installed in a working environment, according to the conditions of the working environment, the generation of the metadata, and has a storage device that holds generation management data including configuration data conditions indicating the configuration of the working environment and event conditions indicating the work content in the working environment. Receives an input of new conditions of the working environment including the current time, the configuration data conditions, and the event conditions. Refers to the generation management data and determines whether there is an existing generation entry of the generation management data including the configuration data condition or the event condition that matches the configuration data condition or the event condition included in the new conditions. When there is an existing generation entry, adds the new conditions of the working environment to the existing generation entry and newly creates an entry of the generation management data related to the metadata added to the sensor data. A data extraction method characterized by the above.

11. On a computer, Stores in a storage device generation management data including configuration data conditions indicating the configuration of a working environment and event conditions indicating the work content in the working environment, and manages, based on a version number given to metadata added to time-series sensor data collected from sensors installed in the working environment, the generation of the metadata. Receives an input of new conditions of the working environment including the current time, the configuration data conditions, and the event conditions. Refers to the generation management data and determines whether there is an existing generation entry of the generation management data including the configuration data condition or the event condition that matches the configuration data condition or the event condition included in the new conditions. When there is an existing generation entry, adds the new conditions of the working environment to the existing generation entry and newly creates an entry of the generation management data related to the metadata added to the sensor data. A data extraction program characterized by causing each process to be executed.

Images