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

Abstract

A knowledge management device is provided with: an acquisition unit that acquires work information of a work process by a worker; a determination unit that determines the work state of the worker on the basis of the work information acquired by the acquisition unit; a question generation unit that generates a question sentence relating to the work state of the worker on the basis of the determination result of the determination unit; a query unit that transmits the question sentence generated by the question generation unit and receives an answer to the question sentence; and an information registration unit that registers knowledge information in the storage unit, the knowledge information being obtained by associating the work process with the answer received by the inquiry unit.

Description

Technical Field

[0001] This disclosure relates to a knowledge management device, a knowledge management method, and a knowledge management program for managing knowledge information about an operator's work. Background Technology

[0002] In the past, insights (knowledge) for work processes were usually transferred and learned through manufacturing instructions, guidance from experienced seniors, or training programs.

[0003] However, tacit insights (knowledge) involve so many aspects that they are difficult to convey in a formal way through individual tasks.

[0004] In this regard, for example, Japanese Patent Application Publication No. 2002-312375 (Patent Document 1) proposes a method for realizing the sharing of tacit knowledge. In addition, Japanese Patent Application Publication No. 2003-196203 (Patent Document 2) proposes a method for collecting information to realize databases.

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: Japanese Patent Application Publication No. 2002-312375

[0008] Patent Document 2: Japanese Patent Application Publication No. 2003-196203 Summary of the Invention

[0009] The problem that the invention aims to solve

[0010] On the other hand, insights (knowledge) regarding work processes are multifaceted, and further improvements are needed in managing them efficiently.

[0011] This disclosure was made to solve the above-mentioned problems, and its purpose is to provide a knowledge management device, knowledge management method and knowledge management program that can efficiently manage knowledge information related to the operator's work procedures.

[0012] Methods for solving problems

[0013] One example of the knowledge management device disclosed herein includes: an acquisition unit that acquires work information of an operator's work procedures; a determination unit that determines the operator's work status based on the work information acquired by the acquisition unit; a question generation unit that generates a question related to the operator's work status based on the determination result of the determination unit; an inquiry unit that sends the question generated by the question generation unit and receives answers to the question; and an information registration unit that registers knowledge information in a storage unit, wherein the knowledge information is obtained by associating the work procedures with the answers received by the inquiry unit. According to this structure, the operator's work status is determined, a question is generated based on the determination result, and the knowledge information obtained by associating the received answers with the work procedures is registered in the storage unit, thus enabling efficient management of knowledge information related to the operator's work procedures.

[0014] The knowledge management device also includes an analysis unit that analyzes the work information acquired by the acquisition unit to generate work status information. A determination unit determines the worker's work status based on a comparison between the work status information generated by the analysis unit and indicator benchmark information representing the worker's work procedures. Because this structure determines the worker's work status based on a comparison between the work status information generated by the analysis unit and the indicator benchmark information, the determination can be performed using a simple method.

[0015] The knowledge management device also includes an indicator benchmark registration unit for registering indicator benchmark information. This structure enables the registration of indicator benchmark information, thus allowing for efficient management of knowledge information related to the operator's work procedures.

[0016] The knowledge management device also includes a description information registration unit, which registers description information related to work procedures. A question generation unit generates questions related to the operator's work status based on the judgment results of the judgment unit and the description information. This structure allows for the generation of questions based on description information, thus increasing the flexibility of question generation and efficiently managing knowledge information related to the operator's work procedures.

[0017] The knowledge management device also includes an information prompting unit that prompts the operator with the knowledge information registered in the storage unit. This structure enables the operator to be prompted with knowledge information, thus facilitating improvements to the operator's work processes.

[0018] The knowledge management device also includes a personnel registration section, which registers information about personnel related to the operator's work processes. An inquiry section sends questions based on this personnel information. This structure allows for the registration of personnel information, thus enabling questions to be sent to the appropriate personnel in a simple manner.

[0019] The knowledge management device also includes an organization section that determines whether to register the responses received by the inquiry section as new knowledge information. Based on the organization section's determination, the information registration section registers the knowledge information in the storage section. This knowledge information is obtained by associating work processes with the responses received by the inquiry section. According to this structure, the organization section determines whether to register new knowledge information, thus enabling efficient management of appropriate knowledge information.

[0020] The processing department performs a similarity assessment on the responses received by the inquiry department and the knowledge information registered in the storage department, following natural language processing principles. Based on the similarity assessment result, it determines whether to register the information as new knowledge. According to this structure, by following natural language processing principles for similarity assessment, it is easy to determine whether to register new knowledge information based on the similarity assessment result, thus enabling efficient management of knowledge information related to the operator's work processes.

[0021] The knowledge management method disclosed herein includes the following steps: obtaining work information of an operator's work procedures; determining the operator's work status based on the obtained work information; generating a question related to the operator's work status based on the determination result; sending the generated question; receiving answers to the question; and registering knowledge information in a storage unit, wherein the knowledge information is obtained by associating the work procedures with the received answers. According to this structure, determining the operator's work status, generating a question based on the determination result, and registering the knowledge information obtained by associating the received answers with the work procedures in the storage unit, it is possible to efficiently manage knowledge information related to the operator's work procedures.

[0022] This disclosed knowledge management program is a knowledge management program that enables a computer to execute a knowledge management method, which includes the following steps: obtaining work information of an operator's work procedures; determining the operator's work status based on the obtained work information; generating a question related to the operator's work status based on the determination result; sending the generated question; receiving an answer to the question; and registering knowledge information in a storage unit, wherein the knowledge information is obtained by associating the work procedures with the received answers. According to this structure, by determining the operator's work status, generating a question based on the determination result, and registering the knowledge information obtained by associating the received answers with the work procedures in the storage unit, knowledge information related to the operator's work procedures can be managed efficiently.

[0023] The effects of the invention

[0024] The knowledge management device, knowledge management method, and knowledge management procedure disclosed herein can efficiently manage knowledge information related to the operator's work procedures. Attached Figure Description

[0025] Figure 1 This is a diagram illustrating the structure of the knowledge management system 1 in the implementation method.

[0026] Figure 2 This is a diagram illustrating the structure of server 20 in the implementation method.

[0027] Figure 3 This diagram illustrates the functional blocks of server 20 in the implementation method.

[0028] Figure 4 This is a flowchart illustrating the processing of server 20 in the implementation method.

[0029] Figure 5 This is a subroutine flowchart that explains the determination process of the determination unit 115 in the implementation method.

[0030] Figure 6 This diagram illustrates the personnel involved in the implementation method.

[0031] Figure 7 This is a diagram illustrating the explanatory information table for the implementation method.

[0032] Figure 8 This is a diagram illustrating the indicator benchmark table for the implementation method.

[0033] Figure 9 This diagram illustrates the determination process of the determination unit 115 in the implementation method.

[0034] Figure 10 This is a diagram illustrating the work data table as work status information in the implementation method.

[0035] Figure 11 This is a diagram illustrating the knowledge data table for the implementation method.

[0036] Figure 12 This is a diagram illustrating the query screen based on the query unit 117 in the implementation method.

[0037] Figure 13 This diagram illustrates the knowledge information prompting based on the information prompting unit 118 in the implementation method.

[0038] Figure 14 This is a diagram illustrating another query screen based on the query unit 117 in the implementation method.

[0039] Figure 15 This is another figure illustrating the knowledge information prompting based on the information prompting unit 118 in the implementation method.

[0040] Figure 16 This diagram illustrates the processing of the finishing section 120 in the implementation method. Detailed Implementation

[0041] Embodiments of the present invention will be described in detail with reference to the accompanying drawings. Furthermore, identical or corresponding parts in the drawings will be labeled with the same reference numerals, and their descriptions will not be repeated.

[0042] (Overall structure)

[0043] The structure of the knowledge management system 1 in the implementation method will be described. Figure 1 This is a diagram illustrating the structure of the knowledge management system 1 in the implementation method. (Refer to...) Figure 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 camera 30 are connected to the server 20. As an example, it shows a scenario where an operator uses a workpiece W configured on a workbench TB to perform a product assembly process. As an example, the assembly process involves assembling the product in the sequence of slider installation, screw tightening, and cover installation. As an example, the operator uses a screwdriver DR to perform the screw tightening operation. The scenario where the operator repeatedly performs the product assembly process is explained.

[0044] As an example, projector 10 projects (outputs) information related to the specified task onto the projection area DP, which is part of the workbench TB. Camera 30 captures the operator's specified task status and outputs the captured image data as task information to server 20.

[0045] Server 20 acquires image data captured by camera 30 as work information, determines the operator's work status based on the acquired work information, and generates a question related to the operator's work status based on the determination result. In addition, server 20 sends the generated question, receives the answer to the question, and registers the knowledge information obtained by associating the work process with the received answer in the storage unit.

[0046] This example illustrates the situation where an operator performs an assembly process, but it is not limited to this; other processes could also be performed. For example, it could also be the situation where an operator performs a hole-making process or a conveying process.

[0047] In this example, we will describe the case where projector 10 is used to project (output) the work instructions for a specified task onto the projection area DP. However, this is not the only example; for instance, a display device can be used instead of projector 10 to display the work instructions for the specified task to the operator. Furthermore, in this example, we will describe the case where the work instructions for the specified task are displayed visually. However, for example, a speaker can also be used to display the information to the operator audibly. Alternatively, a speaker and projector 10 can be used in combination.

[0048] Terminals 50, 52, and 54 are connected to server 20 via a network. Terminals 50, 52, and 54 are configured to answer inquiries from server 20 regarding the operator's work status. Terminals 50, 52, and 54 correspond to the relevant personnel MA, MB, and MC, respectively, as described later.

[0049] Furthermore, this example describes the case where 3 terminals are connected to server 20, but it can also be configured to connect to more terminals, or even to connect to only 1 terminal.

[0050] Figure 2 This is a diagram illustrating the structure of server 20 in the implementation method. (Refer to...) Figure 2The server 20 typically has a structure that follows a general computer architecture. Specifically, the server 20 includes a processor 101 such as a CPU (Central Processing Unit) or MPU (Micro-Processing Unit), memory 102, 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 in a manner that enables data communication. The processor 101 performs various processing tasks in this embodiment by expanding and executing various programs stored in the storage 103 in the storage 102. The storage 102 is typically a volatile storage device such as DRAM (Dynamic Random Access Memory) and stores programs read from the storage 103. The storage 103 is typically a non-volatile magnetic storage device such as a hard disk drive. The storage 103 stores a knowledge management program 131 executed by the processor 101. In addition, the storage 103 stores job information 132, knowledge information 133, indicator benchmark information 134, relevant personnel information 135, and explanatory information 136. Work information 132 is data received via communication interface 106 and transmitted from camera 30, relating to the operator's work status in response to work instructions. Knowledge information 133 is data registered by information registration unit 119, and in the case of receiving an answer to a question, is data related to insights obtained by associating the work procedure with the received answer. Indicator reference information 134 is data registered by indicator reference registration unit 111, and is information representing the indicator references for the operator's work procedures. For example, indicator reference information 134 includes information related to the time of slider installation work corresponding to slider installation. Indicator reference information 134 also includes information related to the time of screw tightening work corresponding to screws, and information related to the time of cover installation work corresponding to covers. Related personnel information 135 is data registered by related personnel registration unit 114, and is information containing the contact information of the personnel who sent the question related to the operator's work procedure. Description information 136 is data registered by description information registration unit 113, and is information used by question generation unit 116 when generating a question.

[0051] Various programs installed in memory 103 are stored in a state similar to those stored on memory cards. Display controller 104 is connected to projector 10 and outputs signals to projector 10 for displaying various information according to internal commands from processor 101. Although not shown, input interface 105 can also mediate data transmission between input devices such as keyboards, mice, touch panels, and dedicated consoles. Communication interface 106 mediates data transmission between processor 101 and external devices (e.g., camera 30). Communication interface 106 typically includes Ethernet (registered trademark), USB (Universal Serial Bus), etc. Furthermore, various programs stored in memory 103 can also be downloaded from distribution servers, etc., via communication interface 106.

[0052] In the case of using a computer constructed according to a general computer architecture as described above, in addition to the application used to provide the functions of this embodiment, an OS (Operating System) can also be installed to provide the basic functions of the computer. In this case, the program of this embodiment can also execute processing by calling necessary modules from the program modules provided as part of the OS in a predetermined order and at predetermined times. That is, the program of this embodiment itself may sometimes not contain the modules described above, but cooperate with the OS to execute processing. Alternatively, some or all of the functions provided by the execution of the knowledge management program 131 may be installed as dedicated hardware circuitry.

[0053] Figure 3 This diagram illustrates the functional blocks of server 20 in the implementation method. (Refer to...) Figure 3 The server 20 constitutes a knowledge management device, including a work information acquisition unit 110, an indicator benchmark registration unit 111, an analysis unit 112, an explanatory information registration unit 113, a relevant personnel registration unit 114, a judgment unit 115, a question generation unit 116, an inquiry unit 117, an information prompting unit 118, an information registration unit 119, and an organization unit 120. Each functional block is implemented by the processor 101 executing the knowledge management program 131.

[0054] The job information acquisition unit 110 acquires job information of the operator's job process. Specifically, the job information acquisition unit 110 acquires image data of the operator's job process from the camera 30 and stores it in the memory 103 as job information 132.

[0055] The indicator reference registration unit 111 registers indicator reference information 134, representing the indicator reference of the operator's work procedure, in the memory 103, for example, according to the input of the manager. In addition, as an example, this example describes the case where the manager registers the indicator reference information 134 in the memory 103 in advance, but it is not limited to this. The information can also be downloaded from an external device connected to the server 20 and registered in the memory 103. The registration method is not particularly limited.

[0056] The analysis unit 112 analyzes the job information 132 stored in the memory 103 to generate job status information for the operator in response to the job instructions. The analysis unit 112 analyzes the image data of the operator's job status input via the camera 30, i.e., the job information 132, and generates the analysis result as job status information.

[0057] The instruction information registration unit 113 registers instruction information 136 related to work indicators and work schedules in the memory 103, for example, according to the manager's input. In addition, this example illustrates the case where the manager registers the instruction information 136 in the memory 103 in advance, but it is not limited to this. The information can also be downloaded from an external device connected to the server 20 and registered in the memory 103. There is no particular limitation on the registration method.

[0058] The determination unit 115 determines the operator's work status based on the work information 132 obtained by the work information acquisition unit 110. Specifically, the determination unit 115 determines the operator's work status based on a comparison between the work status information generated by the analysis unit 112 and the index reference information 134 representing the index reference of the operator's work process.

[0059] The relevant personnel registration unit 114, for example, registers information about relevant personnel related to the operator's work process as relevant personnel information 135 in the memory 103 based on input from the manager. Relevant personnel information 135 is data used to inquire about the reasons for exceeding the target benchmark when the judgment unit 115 determines the operator's work status. Furthermore, this example illustrates the case where the manager pre-registers relevant personnel information 135 in the memory 103, but it is not limited to this. The information can also be downloaded from an external device connected to the server 20 and registered in the memory 103; the registration method is not particularly limited.

[0060] The question generation unit 116 generates a question related to the operator's work status based on the judgment result of the judgment unit 115. Specifically, the question generation unit 116 generates a question related to the operator's work status based on the judgment result of the judgment unit 115 and the explanatory information 136. The inquiry unit 117 sends the question generated by the question generation unit 116 and receives answers to the question. Specifically, the inquiry unit 117 sends the question based on the relevant personnel information 135.

[0061] The processing unit 120 determines whether to register the response received by the inquiry unit 117 as new knowledge information. Specifically, the processing unit 120 performs a similarity determination based on natural language processing on the response received by the inquiry unit 117 and the knowledge information 133 registered in the memory 103, and determines whether to register it as new knowledge information based on the similarity determination result. As an example, the processing unit 120 generates data by converting the response received by the inquiry unit 117 into vector information that follows natural language processing, and determines whether to register the data as new knowledge information by using a similarity determination that uses vector information. In addition, there are various ways to determine similarity, and a method that does not use vector information can also be used. For example, the similarity determination of the response received by the inquiry unit 117 and the knowledge information 133 registered in the memory 103 can be delegated to an external server with a large-scale language model, and the determination result can be obtained.

[0062] The information registration unit 119 registers the knowledge information 133 obtained by associating the work process with the answer received by the inquiry unit 117 in the memory 103. Specifically, the information registration unit 119 registers the knowledge information 133 in the memory 103 based on the determination result of the processing unit 120. The knowledge information 133 is obtained by associating the work process with the answer received by the inquiry unit 117. In this respect, if the information registration unit 119 determines, based on the determination result of the processing unit 120, that the answer received by the inquiry unit 117 is new knowledge information, it will register the knowledge information 133 obtained by associating the work process with the answer received by the inquiry unit 117 in the memory 103. On the other hand, if the information registration unit 119 determines, based on the shaping result of the processing unit 120, that the answer received by the inquiry unit 117 is not new knowledge information, the information registration unit 119 will not register the knowledge information 133 obtained by associating the work process with the answer received by the inquiry unit 117 in the memory 103. As described below, the information registration unit 119 registers knowledge information 133, which includes the converted vector information generated by the processing unit 120, in the memory 103. Furthermore, in this example, as a case study, the following situation will be explained: If the information registration unit 119 receives a new knowledge information response from the inquiry unit 117 as a result of the processing unit 120's determination, it will register the knowledge information 133 obtained by associating the work process with the response received by the inquiry unit 117 in the memory 103. However, it may also register all responses received by the inquiry unit 117 as knowledge information 133 in the memory 103.

[0063] The information prompting unit 118 displays the knowledge information 133 stored in the memory 103 to the operator. As an example, the information prompting unit 118 projects (outputs) the knowledge information 133 onto the projection area DP via the projector 10.

[0064] Figure 4 This is a flowchart illustrating the processing of server 20 in the implementation method. (Refer to...) Figure 4The server 20 obtains the operator's work information (step S2). Specifically, the work information acquisition unit 110 acquires image data of the operator's work process from the camera 30 and stores it in the memory 103 as work information 132. Next, the server 20 analyzes the work information stored in the memory 103 (step S4). The analysis unit 112 analyzes the image data of the operator's work status input via the camera 30, i.e., the work information 132, and generates an analysis result as work status information. For example, the work status information is the time (measurement value) of the operator's work process. For example, if a certain work process includes multiple processes (elements), the time (measurement value) of each process (element) is measured. In addition to the measurement value of the operator's work process, the work status information also includes various information related to the operator's work. Specifically, it includes the operator ID for identifying the operator, the process ID for identifying the work process, and the element ID for identifying the process (element) of the process. The operator ID, process ID, and element ID can be identified through image analysis based on the comparison between the image data of the work status and the reference data used as a comparison benchmark. For example, the operator can be determined by comparing the operator's image, which serves as a comparison benchmark, with the operator's image included in the image data of the work status. The same applies to the process ID and element ID. Furthermore, this example illustrates obtaining the operator ID, stroke ID, and element ID through image analysis, but it is not limited to this; for example, the operator can input the operator ID, stroke ID, and element ID themselves. Regarding the time (measured value) of the operator's work process in this example, as an example, the time from when the operator holds the workpiece W to when it is placed on the worktable TB can be measured by referring to the image data. Alternatively, a sensor used to detect the work can be used, and the measurement time of the sensor can be used. For example, a sensor can be installed in the screwdriver DR, and the screw tightening operation time can be measured based on the time the screwdriver DR is held and the time it is released, according to the sensor's detection signal. Moreover, the above is one example; the measurement time can be obtained in other ways.

[0065] Next, server 20 determines the work status (step S6). Specifically, determination unit 115 determines the work status of the operator based on a comparison between the work status information generated by analysis unit 112 and the index reference information 134 representing the index reference of the operator's work procedure. For example, the case where the index reference of the work procedure is specified within a specified period will be explained. Determination unit 115 compares the time (measured value) of the work procedure with the specified period and determines whether the time (measured value) of the work procedure is included within the specified period. The processing of determination unit 115 will be described later.

[0066] Next, server 20 generates a question based on the determination result (step S8). Specifically, question generation unit 116 generates a question related to the operator's work status based on the determination result of determination unit 115 and the explanatory information 136 stored in memory 103. At this point, question generation unit 116 generates a question if the work status is determined to be an off-standard operation based on the determination result of determination unit 115. On the other hand, question generation unit 116 does not generate a question if the work status is determined to be a normal operation based on the determination result of determination unit 115. In this case, step S18 is entered again, and step S2 is returned if the work has not been completed. If the work status is determined to be an off-standard operation based on the determination result of determination unit 115, and the time (measurement value) of the work process exceeds a specified period, question generation unit 116 generates a question inquiring about an improvement plan for the work. On the other hand, when the question generation unit 116 determines that the work status is an off-standard work as a result of the determination unit 115, and the time (measured value) of the work procedure is less than the specified period, it generates a question to inquire why the work is appropriate and the tricks involved.

[0067] Next, if the server 20 determines that the job status is an off-base job based on the determination result of the determination unit 115, it sends a question (step S10). Specifically, the inquiry unit 117 sends a question generated by the question generation unit 116 based on the relevant personnel information 135.

[0068] Next, server 20 determines whether a response has been received (step S12). Specifically, inquiry unit 117 determines whether a response has been received from the destination of the question. In step S12, server 20 maintains the state of step S12 until a response is received. In step S12, if a response is received ("yes" in step S12), server 20 organizes the response (step S14). Specifically, organization unit 120 determines whether to register it as new knowledge information. Organization unit 120 performs a similarity determination based on the response received by inquiry unit and the knowledge information 133 registered in memory 103, following natural language processing, and determines whether to register it as new knowledge information based on the similarity determination result. Next, server 20 registers the response information as knowledge information (step S16). Specifically, information registration unit 119 registers the knowledge information 133 obtained by associating the work process with the response received by inquiry unit 117 in memory 103. At this point, the information registration unit 119 registers knowledge information 133 in the memory 103 based on the determination result of the processing unit 120. Knowledge information 133 is obtained by associating the work process with the response received by the inquiry unit 117. If the information registration unit 119 determines, based on the determination result of the processing unit 120, that the response received by the inquiry unit 117 is new knowledge information, it registers the knowledge information 133 obtained by associating the work process with the response received by the inquiry unit 117 in the memory 103. If the information registration unit 119 determines, based on the shaping result of the processing unit 120, that the response received by the inquiry unit 117 is not new knowledge information, it does not register the knowledge information 133 obtained by associating the work process with the response received by the inquiry unit 117 in the memory 103. Next, the server 20 determines whether the job has ended (step S18).

[0069] In step S18, if the server 20 determines that the operator has finished the job ("Yes" in step S18), the process ends (ends). For example, the job can also be determined to be finished based on image data from the camera 30 when the operator leaves the workstation TB.

[0070] On the other hand, in step S18, if the server 20 determines that the job has not been completed (in step S18, it is "No"), it returns to step S2 and repeats the above process.

[0071] Figure 5 This is a flowchart illustrating the determination process of the determination unit 115 in the implementation method. (Refer to...) Figure 5The determination unit 115 obtains the index reference information (step S20). Specifically, the determination unit 115 refers to the index reference information 134 stored in the memory 103 to obtain the index reference information corresponding to the work process. For example, in the case of screw tightening operation in the assembly process, information related to the index of the screw tightening operation is obtained. Next, the determination unit 115 determines whether the measured value is within a specified period according to the obtained index reference information (step S22). If the determination unit 115 determines in step S22 that the measured value is within the specified period, it determines that it is a normal operation (step S26). In this case, it proceeds to the next "P". That is, it proceeds to... Figure 4 Step S18. On the other hand, in step S22, if the determination unit 115 determines that the measurement value is not within the specified period, it determines that the operation status is an off-base operation (step S24). Then, the process ends (returns).

[0072] In this case, as an example of work outside the baseline, there are two situations: the time (measurement value) of the work procedure exceeds the specified period and the time (measurement value) of the work procedure is less than the specified period.

[0073] Furthermore, as mentioned above, in the case of work outside the standard, the content of the question differs depending on whether the time (measurement value) of the work procedure exceeds the specified period or whether the time (measurement value) of the work procedure is less than the specified period. Additionally, in the knowledge data table described later, in the case of work outside the standard, the status corresponding to the answer to the question regarding the case where the time (measurement value) of the work procedure exceeds the specified period is recorded as "Bad". In the knowledge data table described later, in the case of work outside the standard, the status corresponding to the answer to the question regarding the case where the time (measurement value) of the work procedure is less than the specified period is recorded as "Good".

[0074] Furthermore, in this example, as a case study, the determination unit 115 has been used to determine whether the time (measured value) of a work process is within a specified period. However, the determination process of the determination unit 115 varies depending on the content of the work process. For example, in the case of a hole-making process, the determination unit 115 may also determine whether the size of the hole is within the specified reference. In addition, in the case of a conveying process, the determination unit 115 may also determine whether the conveying speed is within the specified reference.

[0075] (Registration of relevant personnel information)

[0076] Explanation of the relevant personnel table 135, which represents the relevant personnel information registered by the relevant personnel registration department 114.

[0077] Figure 6 This diagram illustrates the personnel involved in the implementation method. (Refer to...) Figure 6 In the personnel table, each process ID is associated with a corresponding personnel ID. Specifically, process A (assembly) is associated with personnel MA. Process B (hole opening) is associated with personnel MB. Process C (transportation) is associated with personnel MC.

[0078] The inquiry unit 117 refers to the relevant personnel list, obtains the contact information (email address or connection destination information) of the relevant personnel ID associated with the corresponding process ID, and sends a question to the terminal of that relevant personnel ID. For example, a question related to the operator performing process A assembly is sent to the relevant personnel MA. Specifically, the inquiry unit 117 sends the question to the terminal 50 of the relevant personnel MA. The inquiry unit 117 receives a response from the relevant personnel MA via terminal 50. Additionally, a question related to the operator performing process B hole drilling is sent to the relevant personnel MB. Specifically, the inquiry unit 117 sends the question to the terminal 52 of the relevant personnel MB. The inquiry unit 117 receives a response from the relevant personnel MB via terminal 52. Furthermore, a question related to the operator performing process C conveying is sent to the relevant personnel MC. Specifically, the inquiry unit 117 sends the question to the terminal 54 of the relevant personnel MC. The inquiry unit 117 receives a response from the relevant personnel MC via terminal 54. As an example, this shows a scenario where each process ID is associated with the information of one relevant person, but it is not limited to this; multiple relevant persons can also be associated. Furthermore, as an example, this shows a scenario where the contact information differs for each process ID, but the same relevant person can also be associated.

[0079] (Instructions for information registration)

[0080] The description information table representing the description information 136 registered by the description information registration department 113 is described.

[0081] Figure 7 This is a diagram illustrating the explanatory information table for the implementation method. (Refer to...) Figure 7 The explanatory information table is associated with explanatory information corresponding to the process ID. Specifically, corresponding to the assembly process A, explanatory information is associated with the process of assembling the product by processing it in the order of slider installation, screw tightening, and cover installation, as elements representing the process. Similarly, explanatory information is also associated with the opening process B and the conveying process C. The question generation unit 116 generates a question related to the operator's work status based on the determination result of the determination unit 115 and the explanatory information 136 stored in the memory 103.

[0082] (Indicator Benchmark Registration)

[0083] The indicator benchmark table, which represents the indicator benchmark information 134 registered by the indicator benchmark registration department 111, will be explained.

[0084] Figure 8 This is a diagram illustrating the indicator benchmark table for the implementation method. (Refer to...) Figure 8 The benchmark table corresponds to the element ID and is associated with the lower and upper limits of the threshold, as well as the unit. Additionally, it is associated with a "good" indicator. A "good" indicator represents values ​​outside the range of the lower and upper limits of the threshold, indicating which range is considered good.

[0085] As an example, when an indicator is listed as "lower," it indicates that values ​​below the lower limit are considered good. When an indicator is listed as "upper," it indicates that values ​​above the upper limit are considered good.

[0086] Specifically, slider installation A1, screw fastening operation A2, and cover installation A3 are registered as element IDs.

[0087] Corresponding to slider installation A1 are a lower limit of 1, an upper limit of 3, a unit of seconds (sec), and a good indicator "lower". With slider installation A1, the time for operation is shown, with a range of 1 to 3 seconds as the benchmark. A time less than 1 second is better than the benchmark and is considered "Good".

[0088] Corresponding to screw tightening operation A2 are a lower limit of 3, an upper limit of 7, a unit of seconds (sec), and a "good" indicator (lower). In the case of screw tightening operation A2, the time for the operation is shown, with a range of 3 to 7 seconds as the benchmark. If the time is less than 3 seconds, it is better than the benchmark and is considered "Good".

[0089] Corresponding to the A3 cover installation are a lower limit of 1, an upper limit of 5, a unit of seconds (sec), and a "good" indicator (lower). In the case of cover installation A3, the working time is shown, with a range of 1 to 5 seconds as the benchmark. A time less than 1 second is considered better than the benchmark and thus "Good".

[0090] Furthermore, in this example, the cases where slider installation A1, screw fastening operation A2, and cover installation A3 are registered as element IDs are explained, but index benchmarks are also registered for other processes such as hole opening B and conveying C.

[0091] As an example, the determination unit 115 refers to the indicator benchmark table, obtains the indicator benchmark information 134 of the corresponding element ID, and determines whether the measured value is within the specified period. For example, if the measurement value for screw tightening operation A2 is 2 seconds, the determination unit 115 determines it as an off-benchmark operation (Good). For example, if the measurement value for screw tightening operation A2 is 10 seconds, the determination unit 115 determines it as an off-benchmark operation (Bad).

[0092] (Judgment and processing)

[0093] Figure 9 This diagram illustrates the determination process of the determination unit 115 in the embodiment. (Refer to...) Figure 9 In this example, the results of performing screw tightening operation A2 multiple times are shown. This example shows a case where a threshold is set between 3 seconds and 7 seconds. As an example, a value deviating from the threshold (10 seconds) is shown. As an example, if the time (measured value) of this operation deviates from the specified period (10 seconds), the judgment unit 115 determines it as an off-standard operation (Bad), generates a question, and inquires with the relevant personnel MA.

[0094] Figure 10 This is a diagram illustrating the work data table as work status information in the implementation method. (Refer to...) Figure 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 at each time. The example shows the worker ID, process ID, element ID, and measurement value registered in association with the time. Furthermore, image data from the camera 30, i.e., dynamic image data, obtained as work information, can also be registered in association. Moreover, the inquiry unit 117 can also associate the questions generated by the question generation unit 116 with the registered dynamic image data and inquire with relevant personnel. Thus, relevant personnel can easily confirm the work based on the dynamic image data.

[0095] This example illustrates how operator X repeatedly performs the tasks of slider installation A1, screw tightening A2, and cover installation A3 for assembly A. Specifically, it shows the case where operator ID "X", process ID "Assembly A", element ID "Slider Installation A1", and measurement value "4" are associated with the time "2023-11-10T09:11:02". It also shows the case where operator ID "X", process ID "Assembly A", element ID "Screw Tightening A2", and measurement value "10" are associated with the time "2023-11-10T09:11:03". Finally, it shows the case where operator ID "X", process ID "Assembly A", element ID "Cover Installation A3", and measurement value "3" are associated with the time "2023-11-10T09:11:04". The diagram shows the situation where the operator ID "X", process ID "assembly A", element ID "slider installation A1", and measurement value "3" are respectively associated with the time "2023-11-10T09:11:05".

[0096] (Knowledge Information Registration)

[0097] The information registration unit 119 registers the knowledge information 133 obtained by associating the work process with the answer received by the inquiry unit 117 in the memory 103. The diagram illustrates a knowledge data table representing the knowledge information 133 registered by the information registration unit 119.

[0098] Figure 11 This is a diagram illustrating the knowledge data table for the implementation method. (Refer to...) Figure 11 The knowledge data table is associated with the process ID, registering element IDs, operator IDs, statuses, and knowledge points respectively. Specifically, it shows a case where the process ID "Assembly A" is associated with the element ID "Screw Tightening Operation A2", operator ID "X", status "Bad", and knowledge "Please put the screwdriver down straight". It also shows a case where the process ID "Assembly A" is associated with the element ID "Screw Tightening Operation A2", operator ID "X", status "Bad", and knowledge "The screwdriver can be held upright to the left". Finally, it shows a case where the process ID "Assembly A" is associated with the element ID "Screw Tightening Operation A2", operator ID "Y", status "Good", and knowledge "Adopt a posture forming a triangle with the body's center, elbow, and screwdriver".

[0099] In addition, it is not limited to this data table. It can also be used to register related questions, as well as related metadata such as the IDs of relevant personnel who answered the questions, the indicators used for judgment, and product information.

[0100] (Scene of questioning)

[0101] Figure 12 This diagram illustrates the inquiry screen based on the inquiry unit 117 in the embodiment. (Refer to...) Figure 12 This example shows a screen where you can have a chat-like conversation.

[0102] Specifically, the following scenario is illustrated: The inquiry unit 117, in a chat-style dialogue window, inquired with the question, "X spent a lot of time on the screw tightening operation. Do you know why?", on the terminal 50 of the relevant personnel MA. In this example, image data of X's operation was attached to the message containing the question. This image data was obtained via camera 30 as operation information.

[0103] The following scenario is illustrated: Personnel MA, in the chat window of terminal 50, replied to the question with the answer, "You can hold the screwdriver upright to the left." The following scenario is illustrated: The inquiry unit 117 receives the reply from terminal 50 and then replies to the relevant personnel MA's terminal 50 with, "Understood. We will add the suggestion 'You can hold the screwdriver upright to the left.'" Furthermore, this chat-style chat window inquiry is one example; questions can also be asked and answers received via email. Alternatively, it is possible to ask a question without sending a confirmation message from the server to terminal 50.

[0104] Figure 13 This diagram illustrates the knowledge information prompting based on the information prompting unit 118 in the implementation method. (Refer to...) Figure 13 When specified conditions are met, the information prompting unit 118, referring to the knowledge data table, prompts the operator with knowledge information. In this example, the information prompting unit 118 prompts the knowledge information to the projection area DP via the projector 10. Specifically, it shows the case where the message "The screwdriver can be turned to the left" is projected (output) to the projection area DP. Through this display, the operator can confirm the knowledge information output to the projection area DP. For example, when the operator starts the next task, the analysis unit 112 generates task status information. The information prompting unit 118 can also confirm the operator ID, process ID, and element ID contained in the task status information. If the information is consistent, it determines that the specified conditions are met, retrieves the knowledge information registered in the knowledge data table, and prompts it to the projection area DP. In addition, it is not limited to one message. When multiple pieces of knowledge exist, they can all be prompted, or they can be prompted selectively. There are no particular limitations on the prompting method. In this example, the information prompting unit 118 provides knowledge information to the operator by referring to the knowledge data table when the specified conditions are met. However, for example, it could also determine that the specified conditions are met and provide knowledge information to the operator when there is a response from the relevant person MA. The operator can then appropriately improve their work based on this knowledge information.

[0105] Figure 14 This diagram illustrates another inquiry screen based on the inquiry unit 117 in the embodiment. (Refer to...) Figure 14 This example shows a screen where you can have a chat-like conversation.

[0106] Specifically, the following scenario is illustrated: The inquiry unit 117, in a chat-style dialogue window, inquired of the question, "Y is able to complete the screw tightening operation very quickly. Do you know why?", on the terminal 50 of the relevant personnel MA. In this example, the message containing the question is accompanied by image data of Y's operation. This image data was obtained via camera 30 as operation information.

[0107] The following scenario is illustrated: Personnel MA, in the chat window of terminal 50, replied to the question with the response, "Adopt a posture forming a triangle with the center of the body, elbow, and screwdriver." The following scenario is illustrated: The inquiry unit 117 receives the response from terminal 50 and then replies to Personnel MA's terminal 50 with, "Understood. A new suggestion will be added: 'Adopt a posture forming a triangle with the center of the body, elbow, and screwdriver.'" Furthermore, this chat-style chat window inquiry is one example; questions can also be asked and answers received via email. Alternatively, a confirmation message can be sent from the server to terminal 50 without sending a confirmation message.

[0108] Figure 15 This is another figure illustrating the knowledge information prompting based on the information prompting unit 118 in the implementation method. (Refer to...) Figure 15When specified conditions are met, the information prompting unit 118, referring to the knowledge data table, prompts the operator with knowledge information. In this example, the information prompting unit 118 prompts the knowledge information to the projection area DP via the projector 10. Specifically, it shows the message "You can adopt a posture that forms a triangle with your body center, elbow, and screwdriver" being projected (output) to the projection area DP. Through this display, the operator can confirm the knowledge information output to the projection area DP. For example, when the operator starts the next task, the analysis unit 112 generates task status information. The information prompting unit 118 can also confirm the process ID and element ID contained in the task status information, and if the information matches, determine that the specified conditions are met, obtain the knowledge information registered in the knowledge data table, and prompt it to the projection area DP. The operator can use this prompt to appropriately improve the task based on the knowledge information. The knowledge of the "Good" status can be applied by anyone, so it can be prompted regardless of the operator ID, or selectively. There are no particular limitations on the prompting method. In this example, the information prompting unit 118 prompts the operator with knowledge information by referring to the knowledge data table when the specified conditions are met. However, for example, it may also determine that the specified conditions are met and prompt the operator with knowledge information when there is a response from the relevant person MA.

[0109] When providing knowledge information to operators, the information prompting unit 118 can directly display the knowledge message, or it can process and edit the content of the message before displaying it, so that the meaning is conveyed naturally. For example, it can also... Figure 15 You can add messages as shown, or you can change the content of the message to "Let's form a triangle with the center of the body, the elbow, and the screwdriver".

[0110] Figure 16 This diagram illustrates the processing of the finishing section 120 in the implementation method. (Refer to...) Figure 16 The processing unit 120 generates data by converting the responses, i.e., messages, received by the inquiry unit 117 into vector information that follows natural language processing.

[0111] In this example, as a case study, the following situation is shown: for the statement "the screwdriver can be tilted to the left", vector information is generated using an embedding model according to natural language processing. Specifically, the vector information (-0.011 -0.0011 0.032 ……-0.12) is shown.

[0112] In this example, the information registration unit 119 registers knowledge information 133, which contains vector information generated by the sorting unit 120, in the memory 103.

[0113] The sorting department 120 determines whether to register the responses received by the inquiry department 117 as new knowledge information.

[0114] At this point, the processing 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, it compares the vector information of the knowledge portion, and if the similarity is high, it determines that it should not be registered as new knowledge information 133. On the other hand, if the similarity is low, the processing unit 120 adds it to the new knowledge information. As an example, the processing unit 120 makes the determination based on the cosine similarity between the vector information.

[0115] Furthermore, vectorized objects are not limited to knowledge; other information can also be used as a set to generate vector information.

[0116] By registering the vectorized vector information as knowledge information 133 in the memory 103, data comparison or retrieval can be performed efficiently.

[0117] Specifically, by retrieving and extracting knowledge information 133 with the same process ID, it is easy to gather data related to the insights of that process ID, such as data that can be applied to the creation of a work manual for that process.

[0118] Furthermore, by collecting registration history related to a specific process from a knowledge data table and displaying the total number of registrations for each month in a time series, it is possible to show the degree of knowledge accumulation for the target process in a time series manner. For example, this degree of accumulation can also be interpreted as the maturity of the process, quantitatively grasping organizational growth.

[0119] Furthermore, by collecting the status "Good" related to a certain process, gathering knowledge information 133 related to that process that is known as know-how, and sharing this knowledge information with various people, it is easy to realize the sharing of useful knowledge information.

[0120] This approach enables the efficient collection and management of process-related knowledge concerning operators' work. Furthermore, the knowledge database allows for objective reference and understanding of knowledge, facilitating smooth communication regardless of whether the operator is experienced or a beginner. Because insights (knowledge) can be accumulated as data, statistical results can be used to quantitatively represent organizational evolution. The ability to collect and reuse business insights (knowledge) also incentivizes creative research across the organization.

[0121] Embodiments of the invention have been described, but should be considered illustrative rather than limiting in all respects. The scope of the invention is set forth in the claims and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0122] Label Explanation

[0123] 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 Department; 111: Indicator Benchmark Registration Department; 112: Analysis Department; 113: Explanation Information Registration Department; 114: Relevant Personnel Registration Department; 115: Judgment Department; 116: Question Generation Department; 117: Inquiry Department; 118: Information Prompt Department; 119: Information Registration Department; 120: Organization Department; 131: Knowledge Management Program; 132: Work Information; 133: Knowledge Information; 134: Indicator Benchmark Information; 135: Relevant Personnel Information; 136: Explanation Information.

Claims

1. A knowledge management device, wherein, This knowledge management device has the following features: The acquisition department acquires work information related to the operator's work procedures; The determination unit determines the operator's work status based on the work information obtained by the acquisition unit; The question generation unit generates questions related to the operator's work status based on the determination result of the determination unit; The inquiry unit sends out questions generated by the question generation unit and receives answers to those questions; as well as The information registration department registers knowledge information in the storage department, which is obtained by associating the work process with the answers received by the inquiry department.

2. The knowledge management device according to claim 1, wherein, The knowledge management device also includes an analysis unit that analyzes the job information obtained by the acquisition unit to generate job status information. The determination unit determines the operator's work status by comparing the work status information generated by the analysis unit with the indicator benchmark information representing the operator's work procedures.

3. The knowledge management device according to claim 2, wherein, The knowledge management device also has an indicator benchmark registration unit, which registers the indicator benchmark information.

4. The knowledge management device according to claim 1, wherein, The knowledge management device also includes a description information registration unit, which registers description information related to the work process. The question generation unit generates a question related to the operator's work status based on the judgment result of the judgment unit and the explanatory information.

5. The knowledge management device according to claim 1, wherein, The knowledge management device also has an information prompting unit that prompts the operator with the knowledge information registered in the storage unit.

6. The knowledge management device according to claim 1, wherein, The knowledge management device also includes a personnel registration department, which registers information about personnel related to the operator's work process. The inquiry department sends the question based on the information of the relevant personnel.

7. The knowledge management device according to claim 1, wherein, The knowledge management device also includes an organization unit that determines whether to register the responses received by the inquiry unit as new knowledge information. The information registration department registers the knowledge information to the storage department based on the judgment result of the sorting department. The knowledge information is obtained by associating the work process with the answer received by the inquiry department.

8. The knowledge management device according to claim 7, wherein, The sorting unit performs a similarity determination on the answers received by the inquiry unit and the knowledge information registered in the storage unit in accordance with natural language processing, and determines whether to register it as new knowledge information based on the similarity determination result.

9. A knowledge management method, wherein, This knowledge management method has the following steps: Obtain the work information of the operator's work procedures; Based on the obtained work information, determine the work status of the worker; Based on the judgment result, generate a question related to the operator's work status; Send the generated question; Receive the answer to this question; as well as The knowledge information is registered in the storage department. The knowledge information is obtained by associating the work process with the received response.

10. A knowledge management program that causes a computer to execute a knowledge management method, wherein, This knowledge management method has the following steps: Obtain the work information of the operator's work procedures; Based on the obtained work information, determine the work status of the worker; Based on the judgment result, generate a question related to the operator's work status; Send the generated question; Receive the answer to this question; as well as The knowledge information is registered in the storage department. The knowledge information is obtained by associating the work process with the received response.

Images