Knowledge management device, knowledge management method, and knowledge management program

Abstract

This knowledge management device comprises: an acquisition unit which acquires work information of a work process of a worker; a determination unit which determines the work state of the worker on the basis of the work information acquired by the acquisition unit; a question generation unit which generates a question sentence relating to the work state of the worker on the basis of the determination result of the determination unit; an inquiry unit which transmits the question sentence generated by the question generation unit and receives an answer to the question sentence; and an information registration unit which registers, in a storage unit, knowledge information in which the work process and the answer received by the inquiry unit are associated with each other.

Description

Knowledge management device, knowledge management method, and knowledge management program

[0001] The present disclosure relates to a knowledge management device, a knowledge management method, and a knowledge management program for managing knowledge information on work performed by workers.

[0002] In the past, knowledge was generally transferred and learned through manufacturing instructions for work processes, guidance from experienced predecessors, or through training and other development mechanisms.

[0003] However, tacit knowledge is extremely diverse, and it has been difficult to put it into shape and communicate it one by one through manual work.

[0004] In this regard, for example, Japanese Patent Laid-Open Publication No. 2002-312375 (Patent Document 1) proposes a method for sharing tacit knowledge, and Japanese Patent Laid-Open Publication No. 2003-196203 (Patent Document 2) proposes a method for collecting information and creating a database.

[0005] JP 2002-312375 A JP 2003-196203 A

[0006] On the other hand, knowledge about work processes is diverse, and further improvements are needed in terms of efficient management.

[0007] The present disclosure is intended to solve the above-mentioned problems, and aims to provide a knowledge management device, a knowledge management method, and a knowledge management program that are capable of efficiently managing knowledge information related to workers' work processes.

[0008] A knowledge management device according to an example of the present disclosure includes an acquisition unit that acquires work information on a work process of a worker, a determination unit that determines the work status of the worker based on the work information acquired by the acquisition unit, a question generation unit that generates a question regarding the work status of the worker based on a determination result of the determination unit, an inquiry unit that sends the question generated by the question generation unit and receives an answer to the question, and an information registration unit that registers knowledge information that associates the work process with the answer received by the inquiry unit in a storage unit.With this configuration, the work status of the worker is determined, a question is generated based on the determination result, and knowledge information that associates the received answer with the work process is registered in the storage unit, making it possible to efficiently manage knowledge information regarding the work process of the worker.

[0009] The knowledge management device further includes an analysis unit that analyzes the work information acquired by the acquisition unit and generates work status information. The determination unit determines the work status of the worker based on a comparison between the work status information generated by the analysis unit and index reference information that indicates index standards for the worker's work process. This configuration enables a simple determination of the worker's work status based on a comparison between the work status information generated by the analysis unit and the index reference information.

[0010] The knowledge management device further includes an index reference registration unit that registers index reference information. With this configuration, it is possible to register index reference information, and therefore it is possible to efficiently manage knowledge information related to the work processes of workers.

[0011] The knowledge management device further includes an explanatory information registration unit that registers explanatory information related to the work process. The question generation unit generates a question about the work status of the worker based on the determination result of the determination unit and the explanatory information. This configuration allows the question to be generated based on the explanatory information, thereby improving the flexibility of the question and enabling efficient management of knowledge information related to the work process of the worker.

[0012] The knowledge management device further includes an information presentation unit that presents the knowledge information registered in the storage unit to the worker. With this configuration, the knowledge information can be presented to the worker, thereby improving the worker's work process.

[0013] The knowledge management device further includes a related party registration unit that registers information about related parties involved in the worker's work process. The inquiry unit transmits a question based on the related party information. With this configuration, it is possible to register the related party information, so that the question can be transmitted to the appropriate related party in a simple manner.

[0014] The knowledge management device further includes a shaping unit that determines whether to register the answer received by the inquiry unit as new knowledge information. The information registration unit registers knowledge information that associates the work process with the answer received by the inquiry unit based on the determination result of the shaping unit in the storage unit. With this configuration, the shaping unit determines whether to register the knowledge information as new knowledge information, making it possible to efficiently manage appropriate knowledge information.

[0015] The shaping unit performs a similarity determination based on natural language processing on the response received by the query unit and the knowledge information registered in the storage unit, and determines whether to register the information as new knowledge information based on the similarity determination result. With this configuration, it is possible to easily determine whether to register the information as new knowledge information based on the similarity determination result by using similarity determination based on natural language processing, thereby making it possible to efficiently manage knowledge information related to the work processes of workers.

[0016] A knowledge management method according to the present disclosure includes the steps of acquiring work information on a worker's work process, determining the worker's work status based on the acquired work information, generating a question about the worker's work status based on the determination result, sending the generated question, receiving an answer to the question, and registering knowledge information that associates the work process with the received answer in a storage unit.With this configuration, the work status of the worker is determined, a question is generated based on the determination result, and knowledge information that associates the received answer with the work process is registered in the storage unit, making it possible to efficiently manage knowledge information about the worker's work process.

[0017] A knowledge management program according to the present disclosure is a knowledge management program that causes a computer to execute a knowledge management method, and includes the steps of: acquiring work information on a worker's work process; determining the worker's work status based on the acquired work information; generating a question regarding the worker's work status based on the determination result; sending the generated question; receiving an answer to the question; and registering knowledge information that associates the work process with the received answer in a storage unit. With this configuration, the work status of the worker is determined, a question is generated based on the determination result, and knowledge information that associates the received answer with the work process is registered in the storage unit, making it possible to efficiently manage knowledge information regarding the worker's work process.

[0018] According to the knowledge management device, knowledge management method, and knowledge management program of the present disclosure, it is possible to efficiently manage knowledge information related to the work processes of workers.

[0019] 1 is a diagram illustrating a configuration of a knowledge management system 1 according to an embodiment. FIG. 1 is a diagram illustrating a configuration of a server 20 according to an embodiment. FIG. 2 is a diagram illustrating functional blocks of the server 20 according to an embodiment. FIG. 3 is a flow diagram illustrating processing of the server 20 according to an embodiment. FIG. 4 is a subroutine flow diagram illustrating determination processing of a determination unit 115 according to an embodiment. FIG. 5 is a diagram illustrating a related party table according to an embodiment. FIG. 6 is a diagram illustrating an explanation information table according to an embodiment. FIG. 7 is a diagram illustrating an index reference table according to an embodiment. FIG. 8 is a diagram illustrating determination processing of a determination unit 115 according to an embodiment. FIG. 9 is a diagram illustrating a work data table which is work status information according to an embodiment. FIG. 10 is a diagram illustrating a knowledge data table according to an embodiment. FIG. 11 is a diagram illustrating a query screen by a query unit 117 according to an embodiment. FIG. 12 is a diagram illustrating presentation of knowledge information by an information presenting unit 118 according to an embodiment. FIG. 13 is a diagram illustrating another query screen by a query unit 117 according to an embodiment. FIG. 14 is another diagram illustrating presentation of knowledge information by an information presenting unit 118 according to an embodiment. FIG. 15 is a diagram illustrating processing of a shaping unit 120 according to an embodiment.

[0020] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail with reference to the accompanying drawings, in which the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will not be repeated.

[0021] (Overall Configuration) The configuration of a knowledge management system 1 according to an embodiment will be described. FIG. 1 is a diagram illustrating the configuration of the knowledge management system 1 according to an embodiment. Referring to FIG. 1, the knowledge management system 1 includes a projector 10, a server 20, a camera 30, and network-connected terminals 50, 52, and 54. The projector 10 and the camera 30 are each connected to the server 20. As an example, a case is illustrated in which a worker performs a product assembly process using a workpiece W placed on a workbench TB. As an example, the assembly process is a process in which a product is assembled by performing slider installation, screw tightening, and cover installation in that order. As an example, the worker performs screw tightening using a screwdriver DR. A case in which the worker repeatedly performs the product assembly process will be described.

[0022] For example, the projector 10 projects (outputs) information about a predetermined task onto a projection area DP that is a part of the workbench TB. The camera 30 captures an image of the worker performing a predetermined task and outputs the captured image data to the server 20 as task information.

[0023] The server 20 acquires the captured image data input via the camera 30 as work information, determines the work status of the worker based on the acquired work information, and generates a question about the work status of the worker based on the determination result. The server 20 also transmits the generated question, receives answers to the question, and registers knowledge information that associates the work process with the received answer in a storage unit.

[0024] In this example, a case where a worker performs an assembly process will be described as an example, but the present invention is not limited to this and may also be applied to cases where a worker performs other processes, such as a drilling process or a transport process.

[0025] In this example, a case where information on work instructions for a predetermined work is projected (output) onto the projection area DP using the projector 10 will be described as an example, but the present invention is not limited to this. For example, the information on work instructions for a predetermined work may be presented to the worker using a display device instead of the projector 10. Also, in this example, a case where information on work instructions for a predetermined work is presented visually will be described, but the information may also be presented to the worker audio-wise using a speaker, for example. Alternatively, a speaker and the projector 10 may be used in combination.

[0026] The terminals 50, 52, and 54 are connected to the server 20 via a network, and are configured to be able to respond to inquiries about the work status of workers from the server 20. The terminals 50, 52, and 54 are terminals associated with the relevant persons MA, MB, and MC, respectively, which will be described later.

[0027] In this example, a case where three terminals are connected to the server 20 will be described, but the configuration may be such that more terminals are connected, or that only one terminal is connected.

[0028] FIG. 2 is a diagram illustrating the configuration of a server 20 according to an embodiment. Referring to FIG. 2, the server 20 typically has a structure conforming to a general-purpose computer architecture. Specifically, the server 20 includes a processor 101, such as a central processing unit (CPU) or a microprocessing unit (MPU), a memory 102, a storage 103, a display controller 104, an input interface 105, and a communication interface 106. These components are connected to each other via a bus to enable data communication. The processor 101 loads various programs stored in the storage 103 into the memory 102 and executes them to perform various processes according to this embodiment. The memory 102 is typically a volatile storage device such as a dynamic random access memory (DRAM) and stores programs read from the storage 103. The storage 103 is typically a nonvolatile magnetic storage device such as a hard disk drive. The storage 103 stores a knowledge management program 131 executed by the processor 101. The storage 103 also stores work information 132, knowledge information 133, index reference information 134, related party information 135, and explanation information 136. The work information 132 is data related to the work status of the worker in response to the work instructions transmitted from the camera 30 via the communication interface 106. The knowledge information 133 is data registered by the information registration unit 119 and is data related to knowledge that associates the work process with the received answer when a response to a question is received. The index reference information 134 is data registered by the index reference registration unit 111 and is information indicating the index reference for the worker's work process. For example, the index reference information 134 includes information indicating the index reference for the time required for slider installation work corresponding to slider installation. The index reference information 134 includes information indicating the index reference for the time required for screw tightening work corresponding to screws and the time required for cover installation work corresponding to covers. The related party information 135 is data registered by the related party registration unit 114 and is contact information for the related party who sends questions related to the worker's work process.The explanation information 136 is data registered by the explanation information registration unit 113, and is information used by the question generation unit 116 when generating a question.

[0029] The various programs installed in the storage 103 are distributed in a state stored on a memory card or the like. The display controller 104 is connected to the projector 10 and outputs signals to the projector 10 for displaying various information in accordance with internal commands from the processor 101. The input interface 105 may mediate data transmission between input devices such as a keyboard, mouse, touch panel, or dedicated console (not shown). The communication interface 106 mediates data transmission between the processor 101 and an external device (e.g., the camera 30). The communication interface 106 typically includes Ethernet (registered trademark) or USB (Universal Serial Bus). The various programs stored in the storage 103 may be downloaded from a distribution server or the like via the communication interface 106.

[0030] When using a computer having a structure conforming to the above-described general-purpose computer architecture, an operating system (OS) for providing basic computer functions may be installed in addition to an application for providing the functions according to this embodiment. In this case, the program according to this embodiment may execute processing by calling necessary modules provided as part of the OS in a predetermined order and timing. In other words, the program according to this embodiment itself may not include such modules, and may execute processing in cooperation with the OS. Alternatively, some or all of the functions provided by execution of the knowledge management program 131 may be implemented as dedicated hardware circuits.

[0031] 3 is a diagram illustrating functional blocks of server 20 according to an embodiment. Referring to FIG. 3, server 20 constitutes a knowledge management device and includes a task information acquisition unit 110, an index reference registration unit 111, an analysis unit 112, an explanation information registration unit 113, a related person registration unit 114, a determination unit 115, a question generation unit 116, an inquiry unit 117, an information presentation unit 118, an information registration unit 119, and a formatting unit 120. Each functional block is realized by processor 101 executing knowledge management program 131.

[0032] The work information acquisition unit 110 acquires work information on the work process of the worker. Specifically, the work information acquisition unit 110 acquires image data of the work process of the worker from the camera 30 and stores the image data in the storage 103 as work information 132.

[0033] The index standard registration unit 111 registers index standard information 134 indicating index standards for the work process of the worker in the storage 103, for example, according to input by the manager. Note that, in this example, as an example, a case will be described in which the manager registers the index standard information 134 in the storage 103 in advance, but this is not limited thereto, and the information may be downloaded from an external device connected to the server 20 and registered in the storage 103, and there is no particular limitation on the method of registration.

[0034] The analysis unit 112 analyzes the work information 132 stored in the storage 103 based on the work information 132, and generates work status information for the work instruction of the worker. The analysis unit 112 analyzes the work information 132, which is image data of the work status of the worker input via the camera 30, and generates the analysis result as work status information.

[0035] The explanatory information registration unit 113 registers explanatory information 136 relating to task indicators and task processes in the storage 103, for example, according to input from the administrator. Note that, although the present example describes a case in which the administrator registers explanatory information 136 in advance in the storage 103, this is not limiting, and the information may be downloaded from an external device connected to the server 20 and registered in the storage 103, and there are no particular limitations on the method of registration.

[0036] The determination unit 115 determines the work status of the worker based on the work information 132 acquired by the work information acquisition unit 110. Specifically, the determination unit 115 determines the work status of the worker based on a comparison between the work status information generated by the analysis unit 112 and index standard information 134 that indicates index standards for the worker's work process.

[0037] The related party registration unit 114 registers information about related parties involved in the worker's work process as related party information 135 in the storage 103, for example, in accordance with input from the administrator. The related party information 135 is contact data, which is information about related parties used to inquire about the reason why the index standard has been exceeded in accordance with the determination result of the worker's work state by the determination unit 115. Note that, in this example, a case will be described in which the administrator registers the related party information 135 in the storage 103 in advance, but this is not limiting, and the information may be downloaded from an external device connected to the server 20 and registered in the storage 103, and there is no particular limitation on the method of registration.

[0038] The question generation unit 116 generates a question regarding the working status of the worker based on the determination result of the determination unit 115. Specifically, the question generation unit 116 generates a question regarding the working status of the worker based on the determination result of the determination unit 115 and the explanation information 136. The inquiry unit 117 transmits the question generated by the question generation unit 116 and receives an answer to the question. Specifically, the inquiry unit 117 transmits the question based on the related party information 135.

[0039] The shaping unit 120 determines whether to register the answer received by the query unit 117 as new knowledge information. Specifically, the shaping unit 120 performs a similarity determination based on natural language processing between the answer received by the query unit 117 and the knowledge information 133 registered in the storage 103, and determines whether to register the answer as new knowledge information based on the similarity determination result. As an example, the shaping unit 120 generates data by converting the answer received by the query unit 117 into vector information based on natural language processing, and determines whether to register the answer as new knowledge information based on the similarity determination using the vector information. Note that there are various methods for similarity determination, and a method that does not use vector information can also be adopted. For example, an external server having a large-scale language model may be requested to perform a similarity determination between the answer received by the query unit 117 and the knowledge information 133 registered in the storage 103, and the determination result may be acquired.

[0040] The information registration unit 119 registers, in the storage 103, knowledge information 133 that associates a work process with the answer received by the query unit 117. Specifically, the information registration unit 119 registers, in the storage 103, knowledge information 133 that associates a work process with the answer received by the query unit 117 based on the determination result of the shaping unit 120. In this regard, when the shaping unit 120 determines that the answer received by the query unit 117 is new knowledge information as a determination result, the information registration unit 119 registers, in the storage 103, knowledge information 133 that associates a work process with the answer received by the query unit 117. On the other hand, when the shaping unit 120 determines that the answer received by the query unit 117 is not new knowledge information as a result of shaping, the information registration unit 119 does not register, in the storage 103, knowledge information 133 that associates a work process with the answer received by the query unit 117. As will be described later, the information registration unit 119 registers knowledge information 133 including data of the converted vector information generated by the shaping unit 120 in the storage 103. Note that in this example, as an example, when the answer received by the query unit 117 as a determination result of the shaping unit 120 is new knowledge information, the information registration unit 119 registers knowledge information 133 in which the work process is associated with the answer received by the query unit 117 in the storage 103; however, all answers received by the query unit 117 may be registered in the storage 103 as knowledge information 133.

[0041] The information presenting unit 118 presents to the worker the knowledge information 133 stored in the storage 103. The information presenting unit 118 projects (outputs) the knowledge information 133 onto the projection region DP via the projector 10, for example.

[0042] FIG. 4 is a flow diagram illustrating processing by the server 20 according to an embodiment. Referring to FIG. 4, the server 20 acquires work information about the worker (step S2). Specifically, the work information acquisition unit 110 acquires image data of the worker's work process from the camera 30 and stores the image data in the storage 103 as work information 132. Next, the server 20 analyzes the work information stored in the storage 103 (step S4). The analysis unit 112 analyzes the work information 132, which is image data of the worker's work status input via the camera 30, and generates the analysis result as work status information. As an example, the work status information is the time (measured value) of the worker's work process. For example, if a certain work process includes multiple processes (elements), the time (measured value) for each process (element) is measured. Note that the work status information includes various information related to the worker's work in addition to the measured value of the worker's work process. Specifically, the work status information includes a worker ID that identifies the worker, a process ID that identifies the work process, and an element ID that identifies the process (element) of the process. The worker ID, process ID, and element ID can be identified by image analysis based on a comparison between the image data of the work state and reference data serving as a comparison standard. For example, a worker can be identified by comparing the image of the worker that serves as a comparison standard with the image of the worker included in the image data of the work state. The same applies to the process ID and element ID. While this example describes a case in which the worker ID, process ID, and element ID are acquired through image analysis, this is not limited to this. For example, the worker ID, process ID, and element ID may be input by the worker himself / herself. In this example, the time (measured value) of the worker's work process may be measured by, for example, referring to image data, from the time the worker grasps the workpiece W to the time the worker places it on the workbench TB. Alternatively, a sensor for detecting the work may be used to measure the time measured by the sensor. For example, a sensor may be attached to the screwdriver DR, and the work time of the screw tightening work may be measured based on the time the screwdriver DR is grasped and the time the screwdriver DR is released according to the sensor's detection signal. Note that the above is just an example, and the measured time may be acquired using other methods.

[0043] Next, the server 20 determines the work status (step S6). Specifically, the determination unit 115 determines the work status of the worker based on a comparison between the work status information generated by the analysis unit 112 and index standard information 134 indicating the index standard for the worker's work process. For example, a case will be described in which the index standard for the work process is specified to be within a predetermined period. The determination unit 115 compares the time (measured value) of the work process with the predetermined period and determines whether the time (measured value) of the work process falls within the predetermined period. The processing of the determination unit 115 will be described later.

[0044] Next, the server 20 generates a question based on the determination result (step S8). Specifically, the question generator 116 generates a question regarding the worker's work status based on the determination result of the determination unit 115 and the explanation information 136 stored in the storage 103. In this regard, the question generator 116 generates a question when the determination unit 115 determines that the work status is non-standard work. On the other hand, the question generator 116 does not generate a question when the determination unit 115 determines that the work status is normal work. In this case, the process proceeds again to step S18, and if the work has not ended, the process returns to step S2. When the determination unit 115 determines that the work status is non-standard work and the time (measured value) of the work process exceeds a predetermined period, the question generator 116 generates a question asking about an improvement plan for the work. On the other hand, when the judgment unit 115 judges that the work status is non-standard work and the time (measured value) of the work process is less than a predetermined period, the question generation unit 116 generates a question asking for tips on why the work is appropriate.

[0045] Next, when the determination unit 115 determines that the work status is a non-standard work as a result of the determination, the server 20 transmits a question (step S10). Specifically, the inquiry unit 117 transmits the question generated by the question generation unit 116 based on the related party information 135.

[0046] Next, the server 20 determines whether or not an answer has been received (step S12). Specifically, the inquiry unit 117 determines whether or not an answer has been received from the destination of the question. In step S12, the server 20 maintains the state of step S12 until an answer is received. In step S12, if the server 20 receives an answer (YES in step S12), it formats the answer (step S14). Specifically, the formatting unit 120 determines whether or not to register the answer as new knowledge information. The formatting unit 120 performs a similarity determination based on natural language processing between the answer received by the inquiry unit and the knowledge information 133 registered in the storage 103, and determines whether or not to register the answer as new knowledge information based on the similarity determination result. Next, the server 20 registers the answer information as knowledge information (step S16). Specifically, the information registration unit 119 registers, in the storage 103, knowledge information 133 that associates the work process with the answer received by the inquiry unit 117. In this regard, the information registration unit 119 registers, in the storage 103, knowledge information 133 that associates the work process with the answer received by the query unit 117 based on the determination result of the shaping unit 120. When the determination result of the shaping unit 120 indicates that the answer received by the query unit 117 is new knowledge information, the information registration unit 119 registers, in the storage 103, knowledge information 133 that associates the work process with the answer received by the query unit 117. When the determination result of the shaping unit 120 indicates that the answer received by the query unit 117 is not new knowledge information, the information registration unit 119 does not register, in the storage 103, knowledge information 133 that associates the work process with the answer received by the query unit 117. Next, the server 20 determines whether the work has been completed (step S18).

[0047] In step S18, if the server 20 determines that the worker has finished the work (YES in step S18), the server 20 ends the process (END). For example, the server 20 may determine that the work has finished when the worker leaves the workbench TB based on image data from the camera 30.

[0048] On the other hand, in step S18, if the server 20 determines that the work is not completed (NO in step S18), the process returns to step S2 and the above process is repeated.

[0049] FIG. 5 is a subroutine flow diagram illustrating the determination process of the determination unit 115 according to the embodiment. Referring to FIG. 5, the determination unit 115 acquires index reference information (step S20). Specifically, the determination unit 115 acquires index reference information corresponding to the work process by referencing the index reference information 134 stored in the storage 103. For example, when a screw tightening operation is being performed in an assembly process, information related to the indicator of the screw tightening operation is acquired. Next, the determination unit 115 determines whether the measurement value is within a predetermined period in accordance with the acquired index reference information (step S22). If the determination unit 115 determines in step S22 that the measurement value is within the predetermined period, it determines that the work is normal work (step S26). In this case, the process proceeds to the next step "P," i.e., step S18 in FIG. 4. On the other hand, if the determination unit 115 determines in step S22 that the measurement value is not within the predetermined period, it determines that the work status is non-standard work (step S24). The process then ends (returns).

[0050] In this case, there are two cases in which a work is judged to be out of standard: when the time (measured value) of the work process exceeds a predetermined period, and when the time (measured value) of the work process is less than a predetermined period.

[0051] As described above, in the case of non-standard work, the content of the question differs depending on whether the work process time (measured value) exceeds the predetermined period or whether the work process time (measured value) is less than the predetermined period. Also, in the knowledge data table described below, a status ("Bad") is registered in response to a question in the case of non-standard work when the work process time (measured value) exceeds the predetermined period. In the knowledge data table described below, a status ("Good") is registered in response to a question in the case of non-standard work when the work process time (measured value) is less than the predetermined period.

[0052] In this example, the determination unit 115 determines whether the time (measured value) of the work process is within a predetermined period, but the determination unit 115's determination process differs depending on the content of the work process. For example, in the case of a drilling process, the determination unit 115 may determine whether the size of the hole is within a reference index. In the case of a conveying process, the determination unit 115 may determine whether the conveying speed is within a reference index.

[0053] (Registration of Related Person Information) A related person table showing the related person information 135 registered by the related person registration unit 114 will be described.

[0054] 6 is a diagram illustrating a related party table according to an embodiment. Referring to FIG. 6, in the related party table, related party IDs are associated with process IDs. Specifically, related party MA is associated with the assembly A process. Related party MB is associated with the drilling B process. Related party MC is associated with the transport C process.

[0055] The inquiry unit 117 refers to the related party table, acquires contact information (email address or connection information) of the related party ID corresponding to the corresponding process ID, and inquires about a question from the terminal of the related party ID. For example, a question about a worker working in the assembly A process is sent to the related party MA. Specifically, the inquiry unit 117 sends the question to the terminal 50 of the related party MA. The inquiry unit 117 receives a response from the related party MA via the terminal 50. Furthermore, a question about a worker working in the drilling B process is sent to the related party MB. Specifically, the inquiry unit 117 sends the question to the terminal 52 of the related party MB. The inquiry unit 117 receives a response from the related party MB via the terminal 52. Furthermore, a question about a worker working in the transportation C process is sent to the related party MC. Specifically, the inquiry unit 117 sends the question to the terminal 54 of the related party MC. The inquiry unit 117 receives a response from the related party MC via the terminal 54. While an example is shown in which information on one person involved is associated with each process ID, the present invention is not limited to this and information on multiple people involved may be associated. Also, while an example is shown in which contact information differs for each process ID, information on the same person involved may be associated.

[0056] (Explanation Information Registration) The explanation information table showing the explanation information 136 registered by the explanation information registration unit 113 will be described.

[0057] FIG. 7 is a diagram illustrating an explanatory information table according to an embodiment. Referring to FIG. 7, the explanatory information table associates explanatory information with a process ID. Specifically, explanatory information for a process of assembling a product by performing slider attachment, screw tightening, and cover attachment in that order as elements indicating the process progress is associated with the assembly A process. Similarly, explanatory information is associated and registered for the drilling B and conveyance C processes, respectively. The question generator 116 generates a question regarding the worker's work status based on the determination result of the determination unit 115 and the explanatory information 136 stored in the storage 103.

[0058] (Index Reference Registration) An index reference table showing the index reference information 134 registered by the index reference registration unit 111 will be described.

[0059] 8 is a diagram illustrating an index reference table according to an embodiment. Referring to FIG. 8, the index reference table associates lower and upper threshold values ​​and units with element IDs. Also, a goodness index is associated with each element ID. The goodness index indicates which range is good for values ​​outside the range of the lower and upper threshold values.

[0060] For example, if the index "lower" is registered, it indicates that the range of values ​​below the lower limit is good, and if the index "upper" is registered, it indicates that the range of values ​​above the upper limit is good.

[0061] Specifically, slider installation A1, screw tightening A2, and cover installation A3 are registered as element IDs.

[0062] Corresponding to slider attachment A1 are a lower limit value 1, an upper limit value 3, a unit of seconds (sec), and a goodness index lower. In the case of slider attachment A1, the index standard is a range of 1 to 3 seconds as the work time, and a time of less than 1 second indicates that the work is good (good) or better than the standard.

[0063] The screw tightening operation A2 is associated with a lower limit of 3, an upper limit of 7, a unit of seconds (sec), and a goodness index "lower." For the screw tightening operation A2, the index standard is a work time range of 3 to 7 seconds, and a work time of less than 3 seconds is considered to be good (good) above the standard.

[0064] A lower limit value of 1, an upper limit value of 5, a unit of seconds (sec), and a goodness index "lower" are associated with cover installation A3. In the case of cover installation A3, the index standard is a range of 1 to 5 seconds as the work time, and a time of less than 1 second indicates that the work is good (good) or better than the standard.

[0065] In this example, we will explain the case where slider installation A1, screw tightening work A2, and cover installation A3 are registered as element IDs, but index standards are also registered for other processes such as hole drilling B and conveying C.

[0066] For example, the determination unit 115 references the index reference table, acquires the index reference information 134 for the corresponding element ID, and determines whether the measurement value is within a predetermined period. For example, if the measurement value for screw tightening task A2 is 2 seconds, the determination unit 115 determines that the task is outside the standard (Good). For example, if the measurement value for screw tightening task A2 is 10 seconds, the determination unit 115 determines that the task is outside the standard (Bad).

[0067] (Determination Process) FIG. 9 is a diagram illustrating the determination process of the determination unit 115 according to the embodiment. Referring to FIG. 9, in this example, as an example, the results when screw tightening work A2 is performed multiple times are shown. In this example, a case where the threshold is set between 3 seconds and 7 seconds is shown. As an example, a value outside the threshold (10 seconds) is shown. As an example, when the time (measured value) of the work process is a value outside the predetermined period (10 seconds), the determination unit 115 determines that the work is outside the standard (Bad), generates a question, and inquires of the person concerned MA.

[0068] FIG. 10 is a diagram illustrating a work data table, which is work status information according to an embodiment. Referring to FIG. 10 , the analysis unit 112 analyzes the work information 132 and generates the work data table. In this example, work data related to each process is registered for each time period. The table illustrates a case in which a worker ID, a process ID, an element ID, and a measurement value are registered in association with each time period. Image data acquired from the camera 30 as work information, i.e., video data, may also be associated and registered. The inquiry unit 117 may then associate the registered video data with a question generated by the question generation unit 116 and inquire of the relevant person. This allows the relevant person to easily check the work in question according to the video data.

[0069] This example shows a case in which worker X repeatedly performs the slider installation A1, screw tightening A2, and cover installation A3 tasks of assembly A in succession. Specifically, the example shows a case in which worker ID "X," process ID "assembly A," element ID "slider installation A1," and measurement value "4" are associated and registered corresponding to time "2023-11-10T09:11:02." The example shows a case in which worker ID "X," process ID "assembly A," element ID "screw tightening A2," and measurement value "10" are associated and registered corresponding to time "2023-11-10T09:11:03." The example shows a case in which worker ID "X," process ID "assembly A," element ID "cover installation A3," and measurement value "3" are associated and registered corresponding to time "2023-11-10T09:11:04." This shows a case where worker ID "X", process ID "Assembly A", element ID "Slider installation A1", and measurement value "3" are registered in association with the time "2023-11-10T09:11:05".

[0070] (Knowledge Information Registration) The information registration unit 119 registers, in the storage 103, knowledge information 133 that associates a work process with a response received by the inquiry unit 117. FIG. 1 is a diagram illustrating a knowledge data table showing the knowledge information 133 registered by the information registration unit 119.

[0071] FIG. 11 is a diagram illustrating a knowledge data table according to an embodiment. Referring to FIG. 11 , the knowledge data table associates and registers an element ID, a worker ID, a status, and knowledge corresponding to a process ID. Specifically, the table illustrates a case in which an element ID "screw tightening work A2," a worker ID "X," a status "Bad," and knowledge "Please lower the screwdriver straight down" are associated and registered corresponding to the process ID "assembly A." The table illustrates a case in which an element ID "screw tightening work A2," a worker ID "X," a status "Bad," and knowledge "Raise the screwdriver to the left" are associated and registered corresponding to the process ID "assembly A." The table illustrates a case in which an element ID "screw tightening work A2," a worker ID "Y," a status "Good," and knowledge "Adjust your posture so that your body center, elbow, and screwdriver form a triangle" are associated and registered corresponding to the process ID "assembly A."

[0072] In addition, the data table is not limited to this, and question sentences may also be associated and registered, and meta information such as the ID of the person who answered, the index used for the judgment, and product information may also be associated and registered.

[0073] (Inquiry Screen) Fig. 12 is a diagram illustrating an inquiry screen by inquiry unit 117 according to an embodiment. Referring to Fig. 12, in this example, a screen on which a conversation in a chat format can be conducted is shown.

[0074] Specifically, the example shows a case where the inquiry unit 117 makes an inquiry to the terminal 50 of the person involved MA on a chat-style interactive screen, asking the question, "X was taking a long time to tighten screws. Do you know why?" In this example, the example shows a case where image data of X working is attached together with the question message. The image data was acquired as work information via the camera 30.

[0075] The interactive screen of the terminal 50 shows the case where the person involved MA replies to the question, "You should wake the driver to the left." The inquiry unit 117 receives the answer from the terminal 50 and further replies to the terminal 50 of the person involved MA, "I understand. I will add the new advice, 'You should wake the driver to the left.'" Note that the inquiry via the chat-style interactive screen is just one example, and the question content may be inquired by sending and receiving emails, and the answer to the question may be received. Note that it is also possible not to send a confirmation message from the server to the terminal 50.

[0076] FIG. 13 is a diagram illustrating the presentation of knowledge information by the information presenting unit 118 according to an embodiment. Referring to FIG. 13 , the information presenting unit 118 refers to a knowledge data table and presents knowledge information to a worker when a predetermined condition is met. In this example, the information presenting unit 118 presents the knowledge information in the projection area DP via the projector 10. Specifically, a message such as "Please raise the driver to the left" is projected (output) in the projection area DP. This display allows the worker to check the knowledge information output in the projection area DP. For example, when the worker starts the next task, the analysis unit 112 generates work status information. The information presenting unit 118 may check the worker ID, process ID, and element ID included in the work status information, and if the information matches, determine that the predetermined condition is met. The information presenting unit 118 may then acquire the knowledge information registered in the knowledge data table and present it in the projection area DP. Furthermore, if there is more than one knowledge message, all of the knowledge information may be presented, or selectively presented. The presentation method is not particularly limited. In this example, the information presenting unit 118 refers to the knowledge data table and presents knowledge information to the worker when a predetermined condition is met. However, for example, the information presenting unit 118 may determine that a predetermined condition is met when a response is received from the relevant person MA, and present the knowledge information to the worker. The worker can then appropriately improve his or her work based on the presented knowledge information.

[0077] 14 is a diagram illustrating another inquiry screen by inquiry unit 117 according to an embodiment. Referring to FIG. 14, in this example, a screen on which a conversation in a chat format can be conducted is shown.

[0078] Specifically, the example shows a case where the inquiry unit 117 makes an inquiry to the terminal 50 of the person involved MA on a chat-style interactive screen, asking the question, "Y was fastening screws and was able to do it very quickly. Do you know why?" In this example, the example shows a case where image data of Y working is attached together with the question message. The image data is acquired as work information via the camera 30.

[0079] The interactive screen of the terminal 50 shows the case where the person involved MA replies to the question, "Your posture is such that a triangle is formed between the center of your body, your elbow, and the driver." The inquiry unit 117 receives the answer from the terminal 50 and further replies to the terminal 50 of the person involved MA, "I understand. I will add a new piece of advice: 'Your posture is such that a triangle is formed between the center of your body, your elbow, and the driver.'" Note that the inquiry via the chat-style interactive screen is just one example, and the question content may be inquired and the answer to the question may be received by sending and receiving emails. It is also possible not to send a confirmation message from the server to the terminal 50.

[0080] FIG. 15 is another diagram illustrating the presentation of knowledge information by the information presenting unit 118 according to an embodiment. Referring to FIG. 15 , the information presenting unit 118 refers to a knowledge data table and presents knowledge information to a worker when a predetermined condition is met. In this example, the information presenting unit 118 presents the knowledge information in the projection area DP via the projector 10. Specifically, a message such as "It's a good idea to maintain a posture that allows your body center, elbow, and driver to form a triangle" is projected (output) in the projection area DP. This display allows the worker to confirm the knowledge information output in the projection area DP. For example, when the worker starts the next task, the analysis unit 112 generates work status information. The information presenting unit 118 may check the process ID and element ID included in the work status information. If the information matches, the information presenting unit 118 determines that the predetermined condition is met, acquires knowledge information registered in the knowledge data table, and presents it in the projection area DP. The worker can appropriately improve their work based on the presented knowledge information. Since knowledge in the "Good" state is applicable to anyone, it may be presented regardless of the worker ID, or it may be presented selectively. There are no particular limitations on the presentation method. In this example, the information presentation unit 118 refers to the knowledge data table and presents knowledge information to the worker when a predetermined condition is met. However, for example, if there is a response from the related party MA, it may determine that the predetermined condition is met and present the knowledge information to the worker.

[0081] When presenting knowledge information to a worker, the information presenting unit 118 may present the knowledge message as is, or may process and edit the content of the message so that the meaning is conveyed naturally. For example, a message may be added as shown in FIG. 15, or the content of the message may be changed to, for example, "Make a triangle with the center of your body, your elbow, and the screwdriver."

[0082] 16 is a diagram illustrating the processing of the shaping unit 120 according to the embodiment. Referring to FIG. 16, the shaping unit 120 generates data by converting a message, which is a response received by the query unit 117, into vector information that complies with natural language processing.

[0083] In this example, vector information is generated using an embedding model in accordance with natural language processing for the phrase "You should raise the driver to the left." Specifically, the vector information (-0.011, -0.0011, 0.032, ... -0.12) is shown.

[0084] In this example, the information registration unit 119 registers the knowledge information 133 including the vector information generated by the shaping unit 120 in the storage 103 .

[0085] The shaping unit 120 determines whether or not to register the response received by the inquiry unit 117 as new knowledge information.

[0086] At this time, the shaping unit 120 calculates the similarity with data already registered in the knowledge data table and determines whether to register it as new knowledge information 133 based on the similarity. Specifically, the shaping unit 120 compares the vector information of the knowledge portion, and if the similarity is high, it determines not to register it as new knowledge information 133 as similar knowledge information. On the other hand, if the similarity is low, the shaping unit 120 additionally registers it as new knowledge information. As an example, the shaping unit 120 determines the similarity based on the cosine similarity between the vector information.

[0087] The object to be vectorized is not limited to the knowledge portion, and other information may also be included in the set to generate vector information.

[0088] By registering the vectorized vector information in the storage 103 as knowledge information 133, it becomes possible to efficiently compare and search data.

[0089] Specifically, by searching for and extracting knowledge information 133 with the same process ID, it becomes possible to easily aggregate data related to the knowledge of the process ID, which can be used, for example, to create a work manual for the process.

[0090] Furthermore, by collecting the registration history for a certain process from a knowledge data table and displaying the number of registrations per month in chronological order, for example, it is possible to display the degree of knowledge accumulation for the target process in chronological order. For example, this degree of accumulation can be interpreted as the maturity of the process, allowing for a quantitative understanding of organizational growth.

[0091] Furthermore, by collecting the "Good" status for a certain process, knowledge information 133 that serves as so-called tips for the process can be aggregated, and by sharing the knowledge information with various people, it is possible to easily share useful knowledge information.

[0092] This method makes it possible to efficiently collect and manage knowledge about worker work processes. In addition, the knowledge data table makes it possible to objectively refer to and understand knowledge, which can be useful in promoting smooth communication between experts (veterans) and novices alike. Because knowledge can be accumulated as data, the aggregated results can be used to quantitatively express the evolution of an organization. By making it possible to collect and reuse business knowledge, it is also possible to create incentives to encourage originality and ingenuity throughout the organization.

[0093] Although the embodiments of the present invention have been described, the embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, and it is intended to include all modifications within the meaning and scope of the claims.

[0094] 1 Knowledge management system, 10 Projector, 20 Server, 30 Camera, 50, 52, 54 Terminal, 101 Processor, 102 Memory, 103 Storage, 104 Display controller, 105 Input interface, 106 Communication interface, 110 Work information acquisition unit, 111 Index standard registration unit, 112 Analysis unit, 113 Explanation information registration unit, 114 Stakeholder registration unit, 115 Determination unit, 116 Question generation unit, 117 Inquiry unit, 118 Information presentation unit, 119 Information registration unit, 120 Formatting unit, 131 Knowledge management program, 132 Work information, 133 Knowledge information, 134 Index standard information, 135 Stakeholder information, 136 Explanation information.

Claims

1. A knowledge management device comprising: an acquisition unit that acquires work information on a worker's work process; a judgment unit that judges the work status of the worker based on the work information acquired by the acquisition unit; a question generation unit that generates a question regarding the work status of the worker based on the judgment result of the judgment unit; an inquiry unit that transmits the question generated by the question generation unit and receives a reply to the question; and an information registration unit that registers knowledge information in a memory unit that associates the work process with the reply received by the inquiry unit.

2. A knowledge management device as described in claim 1, further comprising an analysis unit that analyzes the work information acquired by the acquisition unit and generates work status information, and the judgment unit judges the work status of the worker based on a comparison of the work status information generated by the analysis unit with index standard information that indicates an index standard for the work process of the worker.

3. The knowledge management device according to claim 2, further comprising an index standard registration section for registering said index standard information.

4. A knowledge management device as described in claim 1, further comprising an explanatory information registration unit that registers explanatory information regarding the work process, wherein the question generation unit generates a question regarding the work status of the worker based on the judgment result of the judgment unit and the explanatory information.

5. The knowledge management device according to claim 1, further comprising an information presentation unit that presents the knowledge information registered in said storage unit to said worker.

6. The knowledge management device according to claim 1, further comprising a related party registration unit that registers information about related parties involved in the work process of the worker, wherein the inquiry unit transmits the question based on the information about the related parties.

7. A knowledge management device as described in claim 1, further comprising a formatting unit that determines whether or not to register the answer received by the inquiry unit as new knowledge information, and the information registration unit registers in a memory unit knowledge information that associates the work process with the answer received by the inquiry unit based on the determination result of the formatting unit.

8. A knowledge management device as described in claim 7, wherein the reformatting unit performs a similarity determination based on natural language processing between the answer received by the inquiry unit and the knowledge information registered in the memory unit, and determines whether or not to register it as new knowledge information based on the similarity determination result.

9. A knowledge management method comprising the steps of: acquiring work information of a worker's work process; judging the work status of the worker based on the acquired work information; generating a question regarding the work status of the worker based on the judgment result; sending the generated question; receiving an answer to the question; and registering knowledge information in a memory unit that associates the work process with the received answer.

10. A knowledge management program that causes a computer to execute a knowledge management method, comprising the steps of: acquiring work information on a worker's work process; judging the work status of the worker based on the acquired work information; generating a question regarding the work status of the worker based on the judgment result; sending the generated question; receiving an answer to the question; and registering knowledge information that associates the work process with the received answer in a memory unit.

Images