Operational training support device, operational training support system, method, and program

The operation training support system efficiently trains new employees by identifying skilled operators, generating procedure videos, and distributing them to training terminals, addressing the challenge of complex procedures and language barriers.

JP7948406B2Active Publication Date: 2026-11-11NEC CORP
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2025500726
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2023-02-14
Filing Date
2024-01-15
Publication Date
2026-11-11
Estimated Expiration
2044-01-15

AI Technical Summary

Technical Problem

New employees find it difficult to master complex operation procedures, and training them is time-consuming, especially in environments with frequent operator changes or when operators speak different languages.

Method used

An operation training support system that collects logs from operators' tasks, uses machine learning to identify skilled and unskilled performers, generates operation procedure videos from skilled logs, and distributes these videos to training terminals for learning.

Benefits of technology

Facilitates efficient training on operation procedures, reducing the time needed to train new employees and accommodating language differences, while improving model accuracy over time through continuous data updates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007948406000001
    Figure 0007948406000001
  • Figure 0007948406000002
    Figure 0007948406000002
  • Figure 0007948406000003
    Figure 0007948406000003
Patent Text Reader

Abstract

Provided is an operation education support device capable of efficiently carrying out education of an operation procedure of work. A log collection means 71 collects, from a terminal at which an operation for execution of the work was carried out by an operator, a log generated in accordance with the operation by the operator. A determination means 72 uses a first model for determining whether or not an operator who was carrying out an operation when a log was generated is a high-performing operator or a non-high-performing operator to determine, for each log collected, whether or not an operator who was carrying out an operation when the log was generated is a high-performing operator or a non-high-performing operator. A video generation means 73 uses a second model for generating an operation procedure video in accordance with the log to generate an operation procedure video in accordance with the log of the operator who has been determined to be a high-performing operator. A distribution means 74 distributes the operation procedure video to an education terminal that is used by an operator for acquiring the operation procedure.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an operation education support device, an operation education support system, an operation education support method, and an operation education support program.

Background Art

[0002] In general, an operator operates a terminal to perform various types of work. Examples of such work include, for example, artificial satellite operation work, work for determining work shifts for retail store employees (hereinafter referred to as work shift determination work), and the like. However, the work is not limited to these examples.

[0003] There are operation procedures for terminal operations to perform work. A skilled operator can perform operations accurately and quickly. On the other hand, unskilled operators and new operators generally cannot perform operations accurately and quickly.

[0004] Incidentally, Patent Document 1 describes a driving training simulator for conducting driving training. Further, Patent Document 2 describes an evaluation device for evaluating signal operations in traffic.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of Invention

Problems to be Solved by the Invention

[0006] If the operation procedures for performing work are complex, it is difficult for new employees to master the operation procedures, and it takes time to educate the new employees. In particular, in departments where operators change frequently, more time is spent on educating new employees.

[0007] Furthermore, if the operator is a foreigner who speaks a language different from the one displayed on the device, training that foreigner will also take time.

[0008] Therefore, the present invention aims to provide an operation training support device, an operation training support system, an operation training support method, and an operation training support program that can efficiently provide training on work operation procedures. [Means for solving the problem]

[0009] The operation training support device according to the present invention is characterized by comprising: log collection means for collecting logs generated in accordance with the operator's operations from a terminal on which an operator performs operations for the performance of a task; determination means for determining whether the operator performing the operation at the time the log was generated is a good or bad operator, using a first model for determining whether the operator performing the operation at the time the log was generated is a good or bad operator; video generation means for generating operation procedure videos corresponding to the logs of operators determined to be good operators, using a second model for generating operation procedure videos corresponding to the logs; and distribution means for distributing operation procedure videos to training terminals used by operators to learn operation procedures.

[0010] The operation training support system according to the present invention comprises the above-mentioned operation training support device and an educational terminal, wherein the educational terminal includes display control means for displaying operation procedure videos distributed by the operation training support device on a display device.

[0011] The operation training support method according to the present invention is characterized in that a computer collects logs generated in response to the operator's operations from terminals where the operator performs operations for the performance of work, determines whether the operator performing the operation at the time the log was generated is a good or bad operator using a first model for determining whether the operator performing the operation at the time the log was generated is a good or bad operator using a second model for generating operation procedure videos corresponding to the logs of operators determined to be good operators, and distributes the operation procedure videos to training terminals in response to requests from training terminals used by operators to learn operation procedures.

[0012] The operation training support program according to the present invention causes a computer to perform the following: a log collection process that collects logs generated in response to the operator's operations from terminals where the operator performed operations for the purpose of performing work; a determination process that determines whether the operator performing the operation at the time the log was generated is a good or bad operator, using a first model for determining whether the operator performing the operation at the time the log was generated is a good or bad operator; a video generation process that generates operation procedure videos corresponding to the logs of operators determined to be good operators, using a second model for generating operation procedure videos corresponding to the logs; and a distribution process that delivers operation procedure videos to educational terminals in response to requests from educational terminals used by operators to learn operation procedures. [Effects of the Invention]

[0013] According to the present invention, training on operational procedures for work can be conducted efficiently. [Brief explanation of the drawing]

[0014] [Figure 1] This is a schematic diagram showing an example of an operation training support system according to an embodiment of the present invention. [Figure 2] This is a block diagram showing an example configuration of an operation training support device according to an embodiment of the present invention. [Figure 3] This is a block diagram showing an example configuration of an educational terminal. [Figure 4] This is a flowchart showing an example of the processing progress when the operation education support device generates an operation procedure video. [Figure 5] This is a flowchart showing an example of the processing progress when the operation education support device generates an operation procedure video. [Figure 6] This is a flowchart showing an example of the processing progress when the operation education support device distributes an operation procedure video to an education terminal, and the education terminal displays the operation procedure video. [Figure 7] This is a block diagram showing a configuration example of the operation education support device in a case where an external apparatus generates a first model and a second model. [Figure 8] This is a schematic block diagram showing a configuration example of a computer pertaining to the operation education support device of the present invention. [Figure 9] This is a block diagram showing an overview of the operation education support device of the present invention. [Mode for Carrying Out the Invention]

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0016] Figure 1 is a schematic diagram showing an example of the operation education support system according to the embodiment of the present invention.

[0017] Figure 1 illustrates a case where there are two bases for performing operations and providing education to new employees, but the number of bases may be one, or three or more. Further, the present embodiment exemplifies a case where each base is provided with a terminal 20 on which operations for performing operations are performed, and an education terminal 30 used by an operator to learn operation procedures for performing operations. Further, a case where the education terminal 30 is provided at the education site of each base is exemplified. However, the terminal 20 and the education terminal 30 may be the same terminal.

[0018] The operation education support system 100 of the present embodiment includes an operation education support apparatus 1, terminals 20, and education terminals 30. The operation education support apparatus 1 is communicably connected to each terminal 20 and each education terminal 30 via a communication network 40. The number of terminals 20 and the number of education terminals 30 are not particularly limited.

[0019] The terminal 20 is a terminal on which an operator performs operations for executing a task. When the operator performs operations from the start of the task to the end of the task, the terminal 20 generates a log corresponding to the operations of the operator performed from the start of the task to the end of the task, and stores the log. That is, one log is generated for one task execution.

[0020] Examples of the task include, as described above, artificial satellite operation tasks, work shift determination tasks, and the like, but the tasks are not limited to these.

[0021] Note that one operator is not necessarily assigned to one terminal 20, and a plurality of operators may use one terminal 20 on a rotating basis. However, it is assumed that the operator does not change from the start of the task to the end of the task.

[0022] Hereinafter, an operator who can perform operations accurately and quickly is referred to as an excellent operator, and an operator who is not an excellent operator is referred to as a non-excellent operator.

[0023] The operation education support apparatus 1 collects logs generated according to the operations of the operator from each terminal 20. Then, for each collected log, the operation education support apparatus 1 determines whether the operator who performed the operation at the time of generating the log is an excellent operator or a non-excellent operator. The operation education support apparatus 1 generates an operation procedure moving image corresponding to the log of the operator determined to be an excellent operator, and stores the operation procedure moving image. Then, in response to a request from the education terminal 30, the operation procedure moving image is distributed to the education terminal 30.

[0024] The training terminal 30 is a terminal used by operators (especially inexperienced individuals such as newcomers) to learn the operating procedures necessary for performing their duties. The training terminal 30 requests an operating procedure video from the operation training support device 1 and receives the operating procedure video distributed from the operation training support device 1. The training terminal 30 then displays the operating procedure video on its display device. Inexperienced individuals using the training terminal 30 can learn the operating procedures by, for example, viewing the displayed operating procedure video or by competing with the operating procedure video to perform the operations accurately and quickly.

[0025] Figure 2 is a block diagram showing an example configuration of an operation training support device 1 according to an embodiment of the present invention. The operation training support device 1 comprises a first learning data storage unit 2, a first learning unit 3, a first model storage unit 4, a log collection unit 5, a determination unit 6, a second learning data storage unit 7, a second learning unit 8, a second model storage unit 9, a video generation unit 10, a video storage unit 11, and a distribution unit 12.

[0026] The first learning data storage unit 2 is a memory device that stores a set of pairs of logs generated in response to operations judged to be excellent and the label "excellent," and a set of pairs of logs generated in response to operations judged to be poor and the label "poor." In the initial state, for example, the administrator of the operation training support device 1 prepares the above data and stores it in the first learning data storage unit 2. Hereinafter, the set of pairs of logs generated in response to operations judged to be excellent and the label "excellent," and the set of pairs of logs generated in response to operations judged to be poor and the label "poor," will be referred to as the learning data for generating the first model. The first model is a model that, given a log, determines whether the operator who performed the operation when the log was generated is an excellent or poor operator.

[0027] The first learning unit 3 generates a first model using machine learning with the first learning data for model generation stored in the first learning data storage unit 2. The first learning unit 3 then stores the first model in the first model storage unit 4.

[0028] The first model storage unit 4 is a memory device that stores the first model.

[0029] As described later, new data is added to the training data for generating the first model. The first learning unit 3 uses the training data for generating the first model, to which the new data has been added, to regenerate the first model using machine learning, and overwrites and stores the first model in the first model storage unit 4 with this new first model. As a result, the first model stored in the first model storage unit 4 is updated.

[0030] The log collection unit 5 collects logs generated in response to the operator's operations from each terminal 20 where the operator performs operations for business execution, via the communication network 40 (see Figure 1). The log collection unit 5 collects multiple logs.

[0031] The determination unit 6 uses the first model stored in the first model storage unit 4 to determine, for each collected log, whether the operator performing the operation when the log was generated is a competent or incompetent operator. The determination unit 6 applies the log to the first model to determine, for each log, whether the operator performing the operation when the log was generated is a competent or incompetent operator. The determination unit 6 can be said to be an AI (Artificial Intelligence) that, given a log, determines whether the operator performing the operation when that log was generated is a competent or incompetent operator.

[0032] When the judgment unit 6 makes the above judgment, it adds the log used for the judgment and the judgment result (labeled "excellent" or "poor") as new data to the first learning data for model generation stored in the first learning data storage unit 2. As described above, the first learning unit 3 uses the first learning data for model generation to which the new data has been added to regenerate the first model by machine learning, and overwrites and stores the first model in the first model storage unit 4 with this new first model. Therefore, the longer the operation of the operation education support device 1 continues, the more data the learning data for model generation increases. As a result, the judgment accuracy of the first model improves.

[0033] The first learning unit 3 may periodically (for example, every month) regenerate the first model using machine learning with the training data for generating the first model at that time. Alternatively, the first learning unit 3 may regenerate the first model using machine learning with the training data for generating the first model at that time when a command to regenerate the first model is input from the administrator of the operation education support device 1.

[0034] The second learning data storage unit 7 is a storage device that stores a set of pairs of logs and operation procedure videos representing the operation procedure when the log was generated. For example, the administrator of the operation training support device 1 prepares the above data and stores it in the second learning data storage unit 7 in advance. Hereinafter, the set of pairs of logs and operation procedure videos representing the operation procedure when the log was generated will be referred to as the learning data for generating the second model. The second model is a model that, given a log, generates an operation procedure video corresponding to that log. The logs included in the learning data for generating the second model are not limited to logs generated by high-performing users. The logs included in the learning data for generating the second model include logs generated by high-performing users and logs generated by low-performing users.

[0035] The second learning unit 8 generates a second model using machine learning with the second learning data stored in the second learning data storage unit 7. The second learning unit 8 then stores the second model in the second model storage unit 9.

[0036] The second model storage unit 9 is a memory device that stores the second model.

[0037] The second learning unit 8 generates a second model by machine learning in advance before the operation of the operation training support device 1 begins, and stores the second model in the second model storage unit 9. No new data is added to the learning data for generating the second model, and the second model is never updated.

[0038] The video generation unit 10 uses the second model stored in the second model storage unit 9 to generate operation procedure videos corresponding to the logs of operators judged to be high performers. The video generation unit 10 applies the logs of operators judged to be high performers to the second model to generate operation procedure videos corresponding to those logs. It can be said that the video generation unit 10 is an AI that, when given logs, generates operation procedure videos corresponding to those logs.

[0039] The log collection unit 5 collects multiple logs, and the determination unit 6 determines, for each log, whether the operator performing the operation when the log was generated is a high-performing or low-performing operator. Therefore, it is possible that there are multiple logs from operators determined to be high-performing. In this case, the determination unit 6 selects the log of the most high-performing operator from among the logs of operators determined to be high-performing. Specifically, the determination unit 6 selects the log with the shortest time from start to finish from among the multiple logs corresponding to the logs of operators determined to be high-performing. Logs of operators determined to be high-performing can be said to reflect accurate operating procedures, and the determination unit 6 selects the log with the shortest time from start to finish (i.e., the log corresponding to the fastest operation) from among such multiple logs.

[0040] The video generation unit 10 generates a video of an operation procedure by applying the selected log as described above to the second model.

[0041] Furthermore, if there is only one log for an operator who has been judged to be an excellent performer, the video generation unit 10 generates one operation procedure video by applying that single log to a second model.

[0042] The video generation unit 10 stores the generated operation procedure video in the video storage unit 11.

[0043] The video storage unit 11 is a storage device that stores videos of operating procedures.

[0044] The distribution unit 12 distributes the operation procedure videos stored in the video storage unit 11 to the training terminal 30 (a terminal used by operators to learn the operating procedures necessary to perform their duties) in response to a request from the training terminal 30. The training terminal 30 then displays the operation procedure videos distributed from the distribution unit 12 of the operation training support device 1 on its display device.

[0045] The log collection unit 5 and the distribution unit 12 are implemented, for example, by the CPU (Central Processing Unit) of a computer that operates according to an operation training support program, and the communication interface of that computer. In this case, the CPU reads the operation training support program from a program recording medium such as the computer's program storage device, and operates as the log collection unit 5 and the distribution unit 12 using the communication interface according to that program.

[0046] Furthermore, the first learning unit 3, the determination unit 6, the second learning unit 8, and the video generation unit 10 are implemented, for example, by the CPU of a computer that operates according to an operation training support program. In this case, the CPU reads the operation training support program from a program recording medium such as the computer's program storage device, as described above, and operates as the first learning unit 3, the determination unit 6, the second learning unit 8, and the video generation unit 10 according to that program.

[0047] The first learning data storage unit 2, the first model storage unit 4, the second learning data storage unit 7, the second model storage unit 9, and the video storage unit 11 are implemented by the storage devices provided by the computer.

[0048] Figure 3 is a block diagram showing an example configuration of an educational terminal 30. The educational terminal 30 illustrated in Figure 3 comprises a keyboard 31, a mouse 32, a request unit 33, a receiving unit 34, a video storage unit 35, a display control unit 36, and a display device 37.

[0049] The keyboard 31 and mouse 32 are operating devices used by the operator to perform operations. The educational terminal 30 may also be equipped with operating devices other than the keyboard 31 and mouse 32.

[0050] The request unit 33 transmits a request to the operation training support device 1 for the distribution of an operation procedure video. The distribution unit 12 of the operation training support device 1 distributes the operation procedure video to the training terminal 30 in response to the distribution request.

[0051] The receiving unit 34 receives the operation procedure video distributed by the operation training support device 1. The receiving unit 34 then stores the operation procedure video in the video storage unit 35.

[0052] The video storage unit 35 is a storage device that stores operation procedure videos distributed by the operation training support device 1.

[0053] The display control unit 36 ​​displays the operation procedure video stored in the video storage unit 35 on the display device 37. There are two display modes in which the display control unit 36 ​​displays the operation procedure video on the display device 37, for example, as shown below.

[0054] [Display Mode 1] The display control unit 36 ​​simply displays the operation procedure video on the display device 37. In this case, the operator using the training terminal 30 (especially a new employee or other less skilled person) learns the operation procedure necessary to perform the job by viewing the displayed operation procedure video.

[0055] [Display Mode 2] The display control unit 36 ​​displays images corresponding to the actions of the operator of the training terminal 30 (especially inexperienced individuals such as new recruits) alongside a video of the operation procedure on the display device 37. The images corresponding to the operator's actions on the training terminal 30 are images corresponding to the operator's actions simulating the execution of a task. In this case, the operator of the training terminal 30 learns the operation procedure necessary for performing the task by competing to perform accurate and quick actions against the video of the operation procedure.

[0056] The request unit 33 and the receiving unit 34 are implemented, for example, by the CPU of a computer operating according to the educational terminal program, and the communication interface of that computer. In this case, the CPU reads the educational terminal program from a program recording medium such as the computer's program storage device, and operates as the request unit 33 and the receiving unit 34 using the communication interface according to that program.

[0057] The display control unit 36 ​​is implemented, for example, by the CPU of a computer that operates according to a program for the educational terminal. In this case, the CPU reads the program for the educational terminal from a program recording medium such as the computer's program storage device, as described above, and operates as the display control unit 36 ​​according to that program.

[0058] The video storage unit 35 is implemented by a memory device provided by the computer.

[0059] Next, the processing steps of an embodiment of the present invention will be described. Figures 4 and 5 are flowcharts illustrating an example of the processing steps when the operation training support device 1 generates an operation procedure video. Here, it is assumed that the first model has already been generated and stored in the first model storage unit 4, and that the second model has already been generated and stored in the second model storage unit 9. Furthermore, explanations of matters already described will be omitted as appropriate.

[0060] When the operation training support device 1 generates an operation procedure video, first, the log collection unit 5 collects logs from each terminal 20 (see Figure 1) (step S1). In step S1, the log collection unit 5 collects multiple logs.

[0061] Next, the determination unit 6 selects one log from among the multiple logs collected in step S1 (step S2). In step S2, the determination unit 6 selects a log that has not yet been selected. Hereinafter, the log selected in step S2 will be referred to as the selected log.

[0062] Next, the determination unit 6 applies the selection log to the first model to determine whether the operator who performed the operation when generating the selection log is a good or bad operator (step S3).

[0063] Next, the determination unit 6 adds the pair of selection log and determination result (labeled "excellent" or "poor") to the training data for generating the first model (step S4).

[0064] If the judgment result in step S3 is "Excellent Person" (Yes in step S5), the judgment unit 6 stores the selection log in the memory of the operation training support device 1 (not shown in Figure 2) (step S6). After step S6, the process proceeds to step S7. The log stored in memory is the log of the operator who was judged to be an excellent person.

[0065] Furthermore, if the evaluation result in Step S3 is "Non-Performing" (No. in Step S5), Step S6 is skipped and the process proceeds to Step S7.

[0066] In step S7, the determination unit 6 determines whether or not there are any logs among the multiple logs collected in step S1 that were not selected in step S2.

[0067] If there are any logs that were not selected in step S2 (Yes in step S7), the determination unit 6 repeats the operations from step S2 onward.

[0068] If all collected logs have been selected (No. in step S7), proceed to step S8 (see Figure 5).

[0069] In step S8, the determination unit 6 determines whether there is one log or two or more logs stored in memory.

[0070] If there is only one log stored in memory (Yes in step S8), the video generation unit 10 generates an operation procedure video corresponding to that log and stores the operation procedure video in the video storage unit 11 (step S9). The process ends in step S9.

[0071] If there are two or more logs stored in memory (No. in step S8), the determination unit 6 selects one log from among the multiple logs stored in memory (step S10). Specifically, the determination unit 6 selects the log with the shortest time from the start of the operation to the end of the operation from among the multiple logs.

[0072] Following step S10, the video generation unit 10 generates an operation procedure video corresponding to the log selected in step S10 and stores the operation procedure video in the video storage unit 11 (step S11). The process ends in step S11.

[0073] In the above process, new data is added to the training data for generating the first model in step S4. As already explained, the first learning unit 3 uses the training data for generating the first model to which the new data has been added to regenerate the first model by machine learning, and overwrites and stores the first model in the first model storage unit 4 with this new first model. The first learning unit 3 may periodically regenerate the first model by machine learning using the training data for generating the first model at that time. Alternatively, the first learning unit 3 may regenerate the first model by machine learning using the training data for generating the first model at that time when a command to regenerate the first model is input from the administrator of the operation education support device 1. If the operation education support device 1 has overwritten and stored the new first model in the first model storage unit 4, it may execute the operations from step S1 onwards again.

[0074] Figure 6 is a flowchart illustrating an example of the processing flow when the operation training support device 1 distributes an operation procedure video to the training terminal 30, and the training terminal 30 displays the operation procedure video. Explanations of matters already described will be omitted as appropriate.

[0075] When the request unit 33 of the educational terminal 30 receives a transmission command for a distribution request from the operator of the terminal device 30, it sends a request for the distribution of an operation procedure video to the operation training support device 1 (step S21).

[0076] The distribution unit 12 of the operation training support device 1 receives the distribution request. Then, in response to the distribution request, the distribution unit 12 distributes the operation procedure video stored in the video storage unit 11 to the training terminal 30 (step S22).

[0077] The receiving unit 34 of the educational terminal 30 receives the operation procedure video distributed in step S22 and stores the operation procedure video in the video storage unit 35.

[0078] Then, the display control unit 36 ​​displays the operation procedure video on the display device 37 (step S23).

[0079] In step S23, the display control unit 36 ​​may simply display the operation procedure video on the display device 37. In this case, the operator of the training terminal 30 learns the operation procedure necessary to perform the work by viewing the displayed operation procedure video.

[0080] Furthermore, in step S23, the display control unit 36 ​​may display images corresponding to the operation of the training terminal 30 and a video of the operation procedure side by side on the display device 37. In this case, the operator of the training terminal 30 learns the operation procedure necessary for performing the job by competing to perform accurate and quick operations with the video of the operation procedure.

[0081] According to this embodiment, the determination unit 6 determines, for each log, whether the operator performing the operation when the log was generated is a high-performing or low-performing operator. The video generation unit 10 generates an operation procedure video corresponding to the log of the operator determined to be high-performing. This operation procedure video can be said to represent the operation procedure of a high-performing operator, and therefore it can be said to demonstrate accurate and quick operations necessary for performing the job. The distribution unit 12 then distributes the operation procedure video to the training terminal 30, and the display control unit 36 ​​of the training terminal 30 displays the operation procedure video on the display device 37. Therefore, the operator of the training terminal 30 can learn the operation procedure necessary for performing the job by viewing the accurate and quick operations necessary for performing the job. Thus, according to this embodiment, training on the operation procedure of the job can be carried out efficiently.

[0082] In particular, when the display control unit 36 ​​displays images corresponding to the operator's actions on the training terminal 30 alongside a video of the operation procedure on the display device 37, the operator of the training terminal 30 can compete to perform accurate and rapid operations against the video of the operation procedure. By engaging in such competition, the operator can learn the operation procedures necessary for performing their duties more efficiently.

[0083] As described above, this embodiment allows for efficient training on operational procedures. Therefore, it eliminates the need to spend time training new employees on operational procedures for performing their jobs. Furthermore, it also eliminates the need to spend time training new employees in departments where operators change frequently.

[0084] Furthermore, even if the operator is a foreigner who speaks a language different from the one displayed, that foreigner will be able to view videos demonstrating accurate and quick operations (operation procedure videos) necessary for performing the task. Therefore, even if that foreigner does not understand the displayed language, they can efficiently learn the operation procedures necessary for performing the task.

[0085] Furthermore, in this embodiment, the determination unit 6 adds new data to the training data for generating the first learning model. Then, the first learning unit 3 uses the training data for generating the first model to which the new data has been added to regenerate the first model using machine learning. Therefore, the longer the operation of the operation training support device 1 continues, the more the amount of data in the training data for generating the first model increases, and the more the determination accuracy of the first model can be improved.

[0086] Next, a modified example of the above embodiment will be described.

[0087] In the above embodiment, the determination unit 6 adds new data to the training data for generating the first learning model, the first learning unit 3 regenerates the first model, and overwrites and stores the updated first model in the first model storage unit 4, thereby updating the first model. However, such an update of the first model is not required. In this case, the determination unit 6 does not need to add new data to the training data for generating the first learning model. That is, the process in step S4 shown in Figure 4 may be omitted. Furthermore, the first learning unit 3 does not need to regenerate the first model.

[0088] Furthermore, the first and second models may be generated by an external device (not shown) to the operation training support device 1. Figure 7 is a block diagram showing an example configuration of the operation training support device 1 when the external device generates the first and second models. Components similar to those shown in Figure 2 are denoted by the same reference numerals as those shown in Figure 2.

[0089] In this case, the operation training support device 1 comprises a first model storage unit 4, a log collection unit 5, a determination unit 6, a second model storage unit 9, a video generation unit 10, a video storage unit 11, and a distribution unit 12. The first learning data storage unit 2, the first learning unit 3, the second learning data storage unit 7, and the second learning unit 8 shown in Figure 2 are not required.

[0090] In Figure 7, the first model storage unit 4 stores the first model generated by an external device. The second model storage unit 9 in Figure 7 stores the second model generated by an external device. For example, the administrator of the operation training support device 1 can store the first model generated by an external device in the first model storage unit 4 and the second model generated by an external device in the second model storage unit 9. In this case, the determination unit 6 does not need to add new data to the learning data for generating the first learning model. That is, the process in step S4 shown in Figure 4 can be omitted. The first learning unit 3 does not need to regenerate the first model. Other operations are the same as in the above embodiment.

[0091] Furthermore, when the display control unit 36 ​​(see Figure 3) displays the operation procedure video on the display device 37, it may adjust the playback speed of the operation procedure video according to the operator's instructions.

[0092] For example, at normal playback speed, the displayed operation movements may be too fast, making it difficult for the operator to grasp the operation procedure. In such cases, the display control unit 36 ​​may slow down the playback speed of the operation procedure video according to the operator's instructions.

[0093] Furthermore, the display control unit 36 ​​may increase the playback speed of the operation procedure video according to the operator's instructions. For example, if the operator is competing to perform accurate and quick operations against the operation procedure video, the display control unit 36 ​​may increase the playback speed of the operation procedure video according to the operator's instructions. In this case, the operator can become accustomed to faster operations and perform operations more quickly.

[0094] Figure 8 is a schematic block diagram showing an example of the configuration of a computer related to the operation training support device 1 of the present invention. The computer 1000 includes, for example, a CPU 1001, a main memory 1002, an auxiliary memory 1003, an interface 1004, and a communication interface 1005.

[0095] The operation training support device 1 in the embodiment of the present invention is implemented by a computer 1000. The operation of the operation training support device 1 is stored in auxiliary storage device 1003 in the form of a program (operation training support program). The CPU 1001 reads the program from the auxiliary storage device 1003, loads the program into the main memory 1002, and executes the processes described in the above embodiment and its modifications according to the program.

[0096] The auxiliary storage device 1003 is an example of a non-temporary tangible medium. Other examples of non-temporary tangible media include magnetic disks, magneto-optical disks, CD-ROMs (Compact Disk Read Only Memory), DVD-ROMs (Digital Versatile Disk Read Only Memory), and semiconductor memory connected via the interface 1004. Furthermore, when a program is distributed to the computer 1000 via a communication line, the computer 1000 that receives the program may load it into the main memory 1002 and execute the processing described in the above embodiment and its modifications according to that program.

[0097] Furthermore, some or all of each component may be implemented by general-purpose or dedicated circuits, processors, etc., or combinations thereof. These may be comprised of a single chip or multiple chips connected via a bus. Some or all of each component may be implemented by a combination of the aforementioned circuits, etc., and programs.

[0098] If some or all of the components are implemented by multiple information processing devices or circuits, these devices may be centrally located or distributed. For example, the information processing devices or circuits may be implemented in a form where each is connected via a communication network, such as a client-server system or a cloud computing system.

[0099] Next, an overview of the present invention will be described. Figure 9 is a block diagram showing an overview of the operation training support device of the present invention. The operation training support device of the present invention comprises a log collection means 71, a determination means 72, a video generation means 73, and a distribution means 74.

[0100] The log collection means 71 (for example, the log collection unit 5) collects logs generated in response to the operator's operations from terminals where the operator has performed operations for the purpose of performing business.

[0101] The determination means 72 (for example, the determination unit 6) uses a first model for determining whether the operator performing the operation when generating the log is a good or bad operator to determine, for each collected log, whether the operator performing the operation when generating the log is a good or bad operator.

[0102] The video generation means 73 (for example, the video generation unit 10) uses a second model for generating operation procedure videos corresponding to logs to generate operation procedure videos corresponding to the logs of operators who have been determined to be excellent performers.

[0103] The distribution means 74 (for example, the distribution unit 12) distributes operation procedure videos to educational terminals used by operators to learn the operation procedures.

[0104] Such a configuration allows for efficient training on operational procedures.

[0105] The embodiments of the present invention described above and their modifications may also be described as follows, but are not limited to the following.

[0106] (Note 1) A log collection means that collects logs generated in response to the operator's operations from terminals where the operator performed operations for business execution, A determination means for determining whether the operator performing the operation when generating the log is a competent or incompetent operator, using a first model for determining whether the operator performing the operation when generating the log is a competent or incompetent operator, A video generation means that generates operation procedure videos corresponding to the logs of operators determined to be excellent, using a second model for generating operation procedure videos corresponding to the logs of operators. The system includes a distribution means for distributing the aforementioned operation procedure videos to an educational terminal used by the operator to learn the operation procedures. An operational training support device characterized by the following features.

[0107] (Note 2) The first model described above is a model generated by machine learning, using a set of pairs of logs generated in response to operations judged to be excellent and labeled as excellent, and a set of pairs of logs generated in response to operations judged to be poor and labeled as poor, as training data for generating the first model. The operational training support device described in Appendix 1.

[0108] (Note 3) The system comprises a first learning means for generating the first model by machine learning. The operational training support device described in Appendix 2.

[0109] (Note 4) The determination means is The log used for the determination and the determination result are added as new data to the training data for generating the first model. The first learning means described above is: Using the training data for generating the first model to which the aforementioned new data has been added, the first model is regenerated by machine learning. The operational training support device described in Appendix 3.

[0110] (Note 5) The second model is a model that has been pre-generated by machine learning, using a set of pairs of logs and operation procedure videos representing the operation procedures when the logs were generated as training data for generating the second model. The operational training support device described in Appendix 1.

[0111] (Note 6) The system comprises a second learning means for generating the second model by machine learning. The operational training support device described in Appendix 5.

[0112] (Note 7) The determination means is If multiple logs exist for an operator judged to be a high-performing operator, the log with the shortest time from the start to the end of the operation will be selected from among the multiple logs. The aforementioned video generation means is Generate an operation procedure video corresponding to the selected log. An operational training support device as described in any of the appendices 1 through 6.

[0113] (Note 8) Educational devices and The device comprises the operation training support device described in claim 1, The aforementioned educational terminal is Includes a display control means for displaying the operation procedure video distributed by the operation training support device on a display device. A training support system for operations, characterized by the following features.

[0114] (Note 9) The display control means is Images corresponding to the user's actions on the educational terminal and the video of the operation procedure are displayed side by side on the display device. The operational training support system described in Appendix 8.

[0115] (Note 10) Computers From terminals where operations for business execution are performed by the operator, logs generated in response to the operator's operations are collected. Using a first model for determining whether the operator performing the operation during log generation is a competent or incompetent operator, for each collected log, it is determined whether the operator performing the operation during log generation is a competent or incompetent operator. Using a second model for generating operation procedure videos based on logs, operation procedure videos are generated based on the logs of operators judged to be high performers. In response to a request from an educational terminal used by an operator to learn the operating procedures, the operating procedure video is distributed to the educational terminal. A method for supporting operational training, characterized by the following features.

[0116] (Note 11) On the computer, A log collection process that collects logs generated in response to the operator's actions from terminals where operations for business execution were performed by the operator. Using a first model for determining whether the operator performing the operation during log generation is a competent or incompetent operator, a determination process is performed for each collected log to determine whether the operator performing the operation during log generation is a competent or incompetent operator. Using a second model for generating operation procedure videos based on logs, a video generation process is performed to generate operation procedure videos based on the logs of operators determined to be high performers, and Distribution process: In response to a request from an educational terminal used by an operator to learn the operating procedures, the distribution process delivers the video of the operating procedure to the educational terminal. A training and support program to enable the execution of the program.

[0117] Although the present invention has been described above with reference to embodiments, the present invention is not limited to the above embodiments. Various modifications to the structure and details of the present invention can be made, which can be understood by those skilled in the art within the scope of the present invention.

[0118] This application claims priority based on Japanese Patent Application No. 2023-020684, filed on 14 February 2023, and incorporates all of its disclosures herein. Potential for Industrial Use

[0119] The present invention is suitably applied to operation training support devices and operation training support systems that enable operators to acquire operating procedures for performing tasks. [Explanation of symbols]

[0120] 1 Operation training support device 2. First learning data storage unit 3. First Learning Department 4. First Model Memory Unit 5. Log Collection Unit 6 Judgment section 7. Second Learning Data Storage Unit 8. Second Learning Department 9. Second Model Memory Unit 10. Video Generation Section 11. Video Memory Unit 12 Distribution Department 20 devices 30 Educational devices 100 Operation Training Support System

Claims

1. A log collection means that collects logs generated in response to the operator's operations from terminals where the operator performed operations for business execution, A determination means for determining whether the operator performing the operation during log generation is a competent or incompetent operator, using a first model for determining whether the operator performing the operation during log generation is a competent or incompetent operator, for each collected log, A video generation means that generates operation procedure videos corresponding to the logs of operators determined to be excellent, using a second model for generating operation procedure videos corresponding to the logs, The system includes a distribution means for distributing the aforementioned operation procedure videos to an educational terminal used by the operator to learn the operation procedures. An operational training support device characterized by the following features.

2. The first model described above is a model generated by machine learning, using a set of pairs of logs generated in response to operations judged to be excellent and labeled as excellent, and a set of pairs of logs generated in response to operations judged to be poor and labeled as poor, as training data for generating the first model. The operation training support device according to claim 1.

3. The system comprises a first learning means for generating the first model by machine learning. The operation training support device according to claim 2.

4. The determination means is The log used for the determination and the determination result are added as new data to the training data for generating the first model. The first learning means is, Using the training data for generating the first model to which the aforementioned new data has been added, the first model is regenerated by machine learning. The operation training support device according to claim 3.

5. The second model is a model that has been pre-generated by machine learning, using a set of pairs of logs and operation procedure videos representing the operation procedures when the logs were generated as training data for generating the second model. The operation training support device according to claim 1.

6. The system comprises a second learning means for generating the second model by machine learning. Operational training support device according to claim 5

7. The determination means is If multiple logs exist for an operator judged to be a high-performing operator, the log with the shortest time from the start to the end of the operation will be selected from among the multiple logs. The aforementioned video generation means is Generate an operation procedure video corresponding to the selected log. An operation training support device according to any one of claims 1 to 6.

8. Educational devices and The device comprises the operation training support device described in claim 1, The aforementioned educational terminal is Includes a display control means for displaying the operation procedure video distributed by the operation training support device on a display device. A training support system for operations, characterized by the following features.

9. The display control means is Images corresponding to the user's actions on the educational terminal and the video of the operation procedure are displayed side by side on the display device. The operation training support system according to claim 8.

10. Computers From terminals where operations for business execution are performed by the operator, logs generated in response to the operator's operations are collected. Using a first model for determining whether the operator performing the operation during log generation is a competent or incompetent operator, for each collected log, it is determined whether the operator performing the operation during log generation is a competent or incompetent operator. Using a second model for generating operation procedure videos based on logs, operation procedure videos are generated based on the logs of operators who were judged to be high performers. In response to a request from an educational terminal used by an operator to learn the operating procedures, the operating procedure video is distributed to the educational terminal. A method for supporting operational training, characterized by the following features.

11. On the computer, A log collection process that collects logs generated in response to the operator's actions from terminals where operations for business execution were performed by the operator. Using a first model for determining whether the operator performing the operation during log generation is a competent or incompetent operator, a determination process is performed for each collected log to determine whether the operator performing the operation during log generation is a competent or incompetent operator. Using a second model for generating operation procedure videos based on logs, a video generation process is performed to generate operation procedure videos based on the logs of operators judged to be high performers, and Distribution process: In response to a request from an educational terminal used by an operator to learn the operating procedures, the distribution process delivers the video of the operating procedure to the educational terminal. A training and support program to enable the execution of the program.

Citation Information

Patent Citations

  • Safe driving teaching system

    JP2009151187A

  • Signal operation evaluation device and computer program

    JP2014164706A

  • Program, computer, and training data generation support method

    JP2015060432A

  • Work support system and work support program

    JP2019109844A

  • Learning assisting system, learning assisting device, and program

    JP2020144233A