Information processing device, manual evaluation system, and manual evaluation method

The information processing device and manual evaluation system analyze user interaction data to improve instruction manuals by identifying unsuitable work blocks and operations, addressing the challenge of varying user skills and characteristics, resulting in enhanced manual explanations and user performance.

WO2024232350A1PCT designated stage expired Publication Date: 2024-11-14HITACHI SYST LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2024/016932
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-10
Filing Date
2024-05-07
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

Existing systems struggle to effectively evaluate and improve instruction manuals for general users, such as those assembling furniture, due to difficulties in collecting reliable work data and identifying unsuitable operations, especially among users with varying skill levels and characteristics.

Method used

An information processing device and manual evaluation system that analyzes user interaction data to identify unsuitable work blocks and operations, using a terminal to record work history and associate it with work elements, allowing for targeted improvements in manual explanations based on user characteristics.

Benefits of technology

Enables efficient evaluation and improvement of instruction manuals by identifying specific work elements that cause inefficiencies, allowing for tailored explanations that enhance user performance, particularly for general users with diverse skill levels and characteristics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024016932_14112024_PF_FP_ABST
    Figure JP2024016932_14112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention addresses the problem of providing a technology for enhancing efficiency of tasks for improving a manual to be referred to by a general user. An information processing device according to one aspect of the present invention comprises a control unit, a storage unit, an input unit, and an output unit, the control unit having an analysis function. The control unit has a function for providing, to a terminal operated by a user from the output unit, a manual instructing the user to perform prescribed work. The storage unit associates contents of the manual with work blocks, work elements associated with the work blocks, and operations to be executed by the user by using the work elements, and stores a step matadata table in which the work blocks, the work elements, and the operations are mutually related and stored. The input unit acquires, for each of the work blocks, a history of processes detected by the terminal at the time of using the manual by the user, and records the same in the storage unit as a work history table. The analysis function refers to the step metadata table and the work history table and analyzes which of the work elements is a factor of work efficiency reduction in a work block, among the work blocks, that has low work efficiency for the user.
Need to check novelty before this filing date? Find Prior Art

Description

Information processing device, instruction evaluation system, and instruction evaluation method

[0001] The present invention relates to a technology for determining the appropriateness of instructions referenced by users, and in particular to a system that supports the improvement of instructions, such as furniture assembly instructions, that are primarily referenced by general users, by performing analysis processing based on historical data of users' references to instructions on their terminals.

[0002] The technologies described in Patent Documents 1 and 2 are known as technologies for evaluating the appropriateness of explanations regarding a model task that a worker uses as a model based on information collected from the worker's work actions.

[0003] Patent Document 1 describes a technology that determines whether a work manual contains tacit knowledge by comparing the work times of skilled and unskilled workers with photographed data of the work process and detecting work steps where work delays occur or where the actions differ.

[0004] Patent Document 2 describes a technology that displays an assembly scenario that has been filmed in advance according to the worker's level of proficiency, measures the time and shoots video for each process, and when the worker's work efficiency is at its highest, saves the filmed work video as a scenario to be displayed from the next time, and notifies workers whose work efficiency is poor of areas to improve their work.

[0005] JP 2020-004274 A JP 2011-134224 A

[0006] However, the systems described in Patent Documents 1 and 2 are only capable of evaluating manuals that are referenced by workers at production sites where detailed work data can be acquired, who have mastered at least the minimum necessary operations, and whose level of proficiency can be classified by the number of experiences, year of joining the company, etc.

[0007] In other words, the technologies described in Patent Documents 1 and 2 can be applied to improving instructions for people who perform work as part of their job, by allowing them to take videos of the work and input work data as each detailed process is completed, but it is difficult to apply them directly to ordinary users, who vary in ability and experience with individual operations and for whom it is not possible to require the acquisition of work data.

[0008] Therefore, an object of the present invention is to provide a technique for improving the efficiency of work to improve instructions that general users refer to.

[0009] One aspect of the present invention is an information processing device having a control unit, a memory unit, an input unit, and an output unit, wherein the control unit has an analysis function, wherein the control unit has a function of providing instructions instructing a user to perform a specified task from the output unit to a terminal operated by the user, the memory unit associates the contents of the instructions with work blocks, work elements associated with the work blocks, and operations that the user should perform on the work elements, and stores a process metadata table that stores the work blocks, the work elements, and the operations in association with each other, the input unit acquires a history of processing detected by the terminal when the user uses the instructions for each work block and records this in the memory unit as a work history table, and the analysis function refers to the process metadata table and the work history table to analyze which of the work elements is causing the user to experience low work efficiency in a work block where work efficiency is low.

[0010] Another aspect of the present invention is an instruction evaluation system including the information processing device described above and a terminal operated by the user, wherein the terminal displays the instruction data in units of the work blocks on one screen of the display either simultaneously or by scrolling the screen, and the processing history detected by the terminal is information on the timing of screen switching on the display.

[0011] Another aspect of the present invention is an instruction evaluation system including the information processing device described above and a terminal operated by the user, wherein the processing history detected by the terminal is at least one of information on the number of operations on an icon or image displayed on the display of the terminal and history information of the chat function.

[0012] Another aspect of the present invention is an instruction evaluation method executed by an information processing device having a control unit, a memory unit, an input unit, and an output unit, the control unit having an analysis function, the memory unit correlating the contents of an instruction instructing a user to perform a predetermined task with a task block, a task element associated with the task block, and an operation to be performed by the user on the task element, and storing a process metadata table that stores the task block, the task element, and the operation in association with each other, the control unit providing the instruction from the output unit to a terminal operated by the user, the input unit acquiring, for each task block, a history of processing detected by the terminal when the user uses the instruction, and the memory unit storing, for a plurality of users, the history of processing in association with a combination of the user ID and the task block. a third step of recording a work history table storing the reflected numerical values; a fourth step in which the analysis function refers to the work history table, calculates the average value of the numerical values ​​for each work block, and compares the average value with a reference value to extract one or more work blocks of interest from the work blocks; a fifth step in which the analysis function refers to the process metadata table and identifies, as one or more work blocks of interest, operations that are unevenly distributed in the work blocks of interest, from among multiple types of operations included in the instruction; a sixth step in which the analysis function refers to the process metadata table and identifies, from the work blocks of interest, those that are associated with at least one of the work elements of the work blocks of interest; and a seventh step in which the output unit outputs information indicating the identified work elements.

[0013] We can provide technology that streamlines the work of improving instructions that general users refer to.

[0014] 1 is a block diagram showing the overall configuration of an instruction evaluation system. FIG. 1 is a data table showing an example of a BEM table of an embodiment. FIG. 2 is a data table showing an example of a work history table of an embodiment. FIG. 3 is a data table showing an example of a user table of an embodiment. FIG. 4 is a flowchart showing processing in a user system. FIG. 5 is an image showing an example of an instruction displayed on a terminal. FIG. 6 is a flowchart of a matching function. FIG. 7 is a flowchart of an analysis function. FIG. 8 is an image showing an example of an instruction evaluation screen displayed on a PC. FIG. 9 is a flowchart showing an example of statistical processing of work history performed by the analysis function of an embodiment. FIG. 10 is a flowchart showing an example of a data table in which matrix T of an embodiment is recorded. FIG. 11 is a flowchart showing an example of a process for determining weak task blocks performed by the analysis function of an embodiment. FIG. 12 is a flowchart showing an example of a data table in which matrix B of an embodiment is recorded. FIG. 13 is a flowchart showing an example of a process for determining weak operations performed by the analysis function of an embodiment. FIG. 14 is a flowchart showing an example of a data table in which matrix M of an embodiment is recorded. FIG. 15 is a flowchart showing an example of a process for determining weak task elements performed by the analysis function of an embodiment. FIG. 16 is a flowchart showing an example of a data table in which matrix E of an embodiment is recorded.

[0015] The embodiments will be described in detail with reference to the drawings. However, the present invention should not be interpreted as being limited to the description of the embodiments shown below. Those skilled in the art will easily understand that the specific configuration can be changed without departing from the concept or spirit of the present invention.

[0016] In the configurations of the embodiments described below, the same parts or parts having similar functions are denoted by the same reference numerals in different drawings, and redundant explanations may be omitted.

[0017] When there are multiple elements having the same or similar functions, they may be described using the same reference numeral with different subscripts. However, when there is no need to distinguish between multiple elements, the subscripts may be omitted.

[0018] The terms "first," "second," "third," and the like used in this specification are used to identify components and do not necessarily limit the number, order, or content of the components. Furthermore, numbers used to identify components are used in different contexts, and a number used in one context does not necessarily indicate the same configuration in another context. Furthermore, this does not prevent a component identified by a certain number from also serving the function of a component identified by another number.

[0019] In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings etc. may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings etc.

[0020] All publications, patents, and patent applications cited herein are incorporated by reference in their entirety.

[0021] As used herein, elements referred to in the singular are intended to include the plural unless the context clearly indicates otherwise.

[0022] The systems described in Patent Documents 1 and 2 do not solve the following problem that arises when evaluating whether instructions for general users, such as furniture assembly instructions, are useful in supporting general users in their work, i.e., when evaluating the appropriateness of the work support provided by instructions for general users.

[0023] First, it is difficult to determine in detail which tasks were not performed properly based on the task data that can be collected from general users. Since general users are not experts in operations (such as "tightening" a screw or "fixing" a component), it is expected that they will not be able to perform each operation properly. Therefore, one method is to have general users check which operations were not performed properly during the work process, such as assembling furniture. However, if a report is required for each operation that was not performed properly, the flow of the assembly work will be interrupted for the user, which could reduce the efficiency of the assembly work that is the intended goal.

[0024] Even if users are encouraged to record their work in detail, it is not guaranteed that the data will be highly reliable. For example, even if users are required to press the confirmation button after completing each operation, it is likely that they will press the confirmation button all at once when they have completed a task.

[0025] Furthermore, in environments such as homes where it is expected that general user instructions will be referred to, it is not possible to take the approach of installing cameras as in production sites and analyzing work videos to identify poor operations.

[0026] Second, for general users, it is not possible to find attributes that clearly correlate with strengths and weaknesses in tasks, such as year of employment or number of years of experience. For this reason, it is not possible to effectively detect inappropriate parts of instructions by focusing on the work performance of workers who have attributes that make them relatively weak at tasks. On the other hand, if the system does not consider which users are weak at the target task, it may not be possible to detect steps that are not a problem for most users but are inappropriate for some users.

[0027] For example, even if there is a process where elderly people are particularly slow to perform their work because they cannot distinguish small symbols on the screws, it is possible that areas for improvement in that process will be overlooked because the work data of a small number of elderly users does not contribute significantly to the statistics based on the work data of all users.

[0028] The technology described in the embodiments proposes the following solution to the above problem: For example, an assembly task (e.g., assembling a chest) that is the subject of an instruction manual is broken down into an "operation (fastening with screws)," a "work element (fastening component A and component B with screws)" that includes the operation (fastening with screws) and the operation target (component A, component B), and an "work block (making a drawer)" that includes one or more work elements.

[0029] The system then obtains the user's browsing history in units of "task blocks," which allows for the acquisition of reliable behavioral data from general users, identifies "weak task blocks" from the "task blocks" in the browsing history, identifies "weak operations" that are common to multiple "weak task blocks," identifies "weak task elements" that include the "weak operations" from among the "weak task blocks," and notifies the user of the identified "weak task elements" as targets for improvement.

[0030] Furthermore, general users are classified by characteristics (gender, age, native language, etc.) that can be extracted from general users and that may correlate with strengths and weaknesses in tasks, and task elements that they are weak at are identified for each characteristic value (gender: male / female, age: elderly / middle-aged / young, native language: Japanese / non-Japanese, etc.).

[0031] This allows the user to simply input their reference history for each task block, allowing the instruction manual creator to determine whether or not the instruction manual needs to be improved for each task element, facilitating the evaluation of the appropriateness of the instruction manual. Furthermore, it is possible to detect task elements that are inappropriately explained only to users with specific characteristic values, even if the number of such elements is small. Additionally, by revising the explanation for each characteristic value, it is possible to deliver the most appropriate instruction manual to each user. Note that in this specification, the term "instructions" is used to refer not only to documents but also to general content that instructs users on specific tasks, such as images, audio, and video.

[0032] (1. System Configuration) Fig. 1 is a configuration diagram of the entire instruction manual evaluation system of the embodiment. Fig. 1 shows a PC 102, an instruction manual management server 103, and a terminal 113. Here, an assembly instruction manual for furniture such as a chest of drawers is taken as a specific example.

[0033] (1-1. Instructions Management Server) First, the configuration of the instructions management server 103 will be described. The instructions management server 103 is managed, for example, by a business that provides products and their instructions. The instructions management server 103 is generally configured as a business server or cloud.

[0034] The PC 102 operated by the instruction manual creator 101 is a general personal computer equipped with an input section, a display section, a memory section, and a control section, and is connected to the communication section 106 of the instruction manual management server 103 via a network line.

[0035] The instruction manual management server 103 is a server equipped with a memory unit 104, a control unit 105, and a communication unit 106, and is connected to the PC 102 and the communication unit 107 of the user system 112 via a network line. The instruction manual management server 103 also includes a general input device (input unit) and an output device (output unit). The communication unit 107 may be considered as one of the input device (input unit) and the output device (output unit). The PC 102 may also be configured as part of the instruction manual management server 103.

[0036] The memory unit 104 of the instruction manual management server 103 has as data an instruction manual DB 104a, a BEM table 104b, a work history table 104c, a user table 104d, a notification message table 104e, and a match table 104f. The memory unit 104 can be configured by any combination of storage devices such as a magnetic disk device and a semiconductor memory.

[0037] The instruction manual DB 104a stores instructions linked to instruction manual IDs. Here, the instruction manual is content for explaining the work procedure for assembling a product to be assembled (for example, a chest), and is composed of a collection of multiple sets of instructions related to a work block (making a drawer) that includes one or more work elements (fastening component A and component B with screws) that include an operation (fastening with screws) and components to be operated (components A and B).

[0038] The data format of the instruction manual may include text, images, video, and audio. The instruction manual data is created in advance by the instruction manual creator 101 or a system administrator and stored in the storage unit 104. The instruction manual data is transmitted to the terminal 113 as needed and made available for use by the user 114.

[0039] 2 is a data table showing an example of the BEM table 104b. The BEM table 104b is a table created in a one-to-one correspondence with the instruction manual, and defines the association between the work blocks that make up the instruction manual, the work elements that make up the work blocks, and the operations that the work elements include.

[0040] Figure 2 shows data for one instruction manual. One or more task block IDs that identify task blocks are associated with an instruction ID that identifies the instruction. One or more task element IDs that identify task elements are associated with one task block ID. One task element includes a set of any number of task targets and any number of operations. In this example, the data contains standard task time data for the task block, but other data for evaluating task efficiency may also be included instead of or in addition to this.

[0041] The data in the BEM table 104b is created in advance by the instruction manual creator 101 or a system administrator and stored in the storage unit 104. In this case, the definitions of the work block, work element, and operation are arbitrary. In the example of FIG. 2, the work elements include two operation targets and one operation, but the number of operation targets and operations is not limited to this. The number of work elements to be included in one work block is arbitrary, but it is preferable to group content that is easy for the user 114 to separate into one work block. As an example, one or more work elements whose explanations can be displayed on one screen of the terminal 113 can be associated with one work block.

[0042] 3 is a data table showing an example of the work history table 104c. The work history table 104c is created on a one-to-one basis for each instruction manual, and links user IDs with information about the user's work history. The work history is recorded for each work block, and may be any information that can be recorded by the user 114 operating the terminal. The work history table 104c is designed to collect data from a large number of users, and the collected data is subjected to statistical processing.

[0043] In the example of Figure 3, the work histories of users with user IDs "1" to "5" are stored for the instruction manual with instruction manual ID "1." One work history is recorded for each work block, such as "drawer assembly" or "storage section assembly," and in this example, the work time (seconds) is stored. The recording means of the work history table 104c will be described later.

[0044] FIG. 4 is a data table showing an example of the user table 104d. In the user table 104d, a user ID that identifies a user registered in the user system is associated with the value of that user's characteristic (hereinafter referred to as characteristic value). In FIG. 4, "gender," "age group," and "native language" are recorded as examples of characteristics. As examples of characteristic values, "male" and "female" are recorded for "gender." The characteristic value may be a category or a numerical value. Furthermore, the characteristics and characteristic values ​​associated with a user may be anything that can be extracted from general users.

[0045] The characteristic values ​​may be investigated by conducting a questionnaire before or after work, or by using information acquired when the user registers in the user system 112. Alternatively, the user table 104d may be created by a system administrator based on information collected in advance and stored in the storage unit 104. Alternatively, the user 114 may input the information from the terminal 113 when using the instructions in the instruction DB 104a.

[0046] The notification table 104e stores, in association with the instruction manual ID, notification messages that notify the instruction manual creator 101 of processes for which the instruction manual management server 103 has determined that the explanation is inappropriate. The data in the notification table 104e is created by the analysis function 105b and stored in the memory unit 104.

[0047] The match table 104f is a table that uses the product ID and characteristic value as a composite primary key and searches for instruction manual IDs, and is used to classify multiple instructions for one product when they are remade for each characteristic value. The data in the match table 104f is created in advance by the instruction manual creator 101 or the system administrator and stored in the memory unit 104.

[0048] The control unit 105 is hardware that functions as a computer, and includes a CPU (Central Processing Unit), RAM (Random Access Memory), and ROM (Read Only Memory) as processing devices, and controls the operation of the instruction manual management server 103. The control unit 105 controls the execution of the matching function 105a and the analysis function 105b. The matching function 105a and the analysis function 105b are realized by the CPU executing programs stored in, for example, RAM or ROM. In terms of the system configuration, the matching function 105a may be considered as a matching function unit, and the analysis function 105b may be considered as an analysis function unit.

[0049] The control unit 105 or the instruction manual management server 103 may be configured as a standalone computer, or any of the input device, output device, processing device, and storage device may be configured as other computers connected via a network. Furthermore, functions equivalent to those configured by software can also be realized by hardware such as FPGAs (Field Programmable Gate Arrays) and ASICs (Application Specific Integrated Circuits).

[0050] The matching function 105 a is responsible for searching for an appropriate instruction manual from the match table 104 f and transmitting the instruction manual to the user system 112 when a request for an instruction manual is received from the user system 112 .

[0051] When the analysis function 105b receives instructions to be evaluated from the PC 102, it identifies processes that may contain inappropriate instructions, issues a notification to notify them, and sends the analysis results to the PC 102.

[0052] (1-2. User System) Next, we will explain the configuration of the user system 112. As a specific example, the user system 112 is configured with a personal computer or a portable information terminal (terminal 113) used by a general user 114.

[0053] The terminal 113 operated by the user 114 is, for example, a typical smartphone, and is equipped with an input unit (touch panel) 111, a display unit 110, a memory unit 108, a control unit 109, and a communication unit 107, and can be connected to the communication unit 106 of the instruction manual management server 103 via a network line.

[0054] The user system 112 stores an instruction manual DB 108a and a work history table 108b in the memory unit 108. The control unit 109 also has a work history acquisition function 109a. The users 114 are assumed to be general users with various characteristics, for example, from elderly people to children.

[0055] The control unit 109 is hardware that functions as a computer, and includes a CPU, RAM, and ROM, and controls the operation of the user system 112. The control unit 109 controls the execution of a work history acquisition function 109a. The work history acquisition function 109a is realized by the CPU executing a program stored in, for example, RAM or ROM. The program is downloaded from, for example, the instruction manual management server 103.

[0056] The work history acquisition function 109a acquires the work history from the operations performed by the user 114 on the terminal 113 for each work block, and records the work history in a work history table 108b (described later). The data in the work history table 108b is sent to the instruction manual management server 103 as needed, and is collected in a work history table 104c.

[0057] An operation performed on the terminal 113 is, for example, "the user 114 tapping the Next button 304" on the task block screen (described later in FIG. 6), i.e., the timing of switching screens, etc. In this case, the task history acquisition function 109a may record the time from when the explanation screen for one task block starts to be displayed until the Next button 306 is tapped (i.e., equivalent to the task time for the task block) as the task history for that task block.

[0058] The storage unit 108 includes an instruction manual DB 108a and a work history table 108b. The instruction manual DB 108a stores instructions requested by the user 114 and returned from the instruction manual management server 103, linked to a product ID and an instruction manual ID. The work history table 108b records information about the user's reference history to instructions, acquired by the work history acquisition function 109a, for each work block, and records the information in association with the instruction manual ID.

[0059] First, we will explain an example of the series of operations in the user system 112 from when the user 114 requests instructions from the instruction management server 103 to when the work history is sent to the instruction management server 103, and the operations in the matching function 105a from when the request from the user 114 is received to when an appropriate instruction is searched for and a reply is sent to the user 114.

[0060] (2. Operation Description) (2-1. Example of Operation of User System) Fig. 5 is a flowchart showing the processing of the user system 112. First, the user 114 starts the terminal 113 and selects a product for which he / she wishes to view the manual using the input unit 111, and the processing shown in Fig. 5 starts (S201).

[0061] When the process starts, the control unit 109 sends the product ID of the selected product and the user ID of the user acquired from the terminal to the instruction manual management server 103 via the communication unit 107, and requests the instruction manual (S202).

[0062] After S202, the instruction manual management server 103 processes the matching function 105a described below, and returns the appropriate instruction manual from the instruction manual DB 104a to the user system 112 via the communication units 106 and 107.

[0063] The instruction manual sent back to the user system 112 is stored in the instruction manual DB 108a by the control unit 109 (S203). The data in the instruction manual DB 108a is considered to be a partial copy of the data in the instruction manual DB 104a. When the user 114 selects an instruction manual for the product to be assembled from the input unit 111, a new work history table 108b is created and the instruction manual ID and user ID are recorded therein (S204).

[0064] When the user 114 selects start of work from the input unit 111, the first work block screen is displayed on the display unit 110 of the terminal 113 (S206).

[0065] 6 is an example of a task block screen displayed on the display unit 110 of the terminal 113, and includes a task block description 301, a task element description 302 that constitutes the task block, and a next button 304. These contents correspond to the data structure of the BEM table 104b shown in FIG. 2 and are created based on the data in the instruction manual DB 108a.

[0066] Here, the task element description 302 may include an image 303. The image is used to provide supplementary explanation of the operation of the task element, and may include information such as images, videos, and text. A chat icon 305 may also be added to the task block screen, and a chat screen may be displayed when the chat icon is pressed. In this case, the user 114 can use chat to ask the person in charge of receiving questions about the task block regarding any questions that arise.

[0067] When the Next button 304 is pressed (S207), the work history acquisition function 109a acquires the work history for the first work block and saves it in the work history table 108b (S208). The work history acquired here may be anything that can be recorded by the user 114 operating the terminal 113. For example, as described above, the work history may be the work time estimated based on the time from when the work block screen is displayed until the Next button 304 is tapped (completion time). Alternatively, it may be the number of times an image is tapped (number of views). Alternatively, it may be the number of times a video is played repeatedly. Furthermore, if a chat function is added, questions sent by the user may also be acquired as work history.

[0068] After S208, the process returns to S205, and the instruction screen for the next task block is displayed. The processes from S206 to S208 are repeated until the instruction screens for all task blocks that make up the instruction manual have been displayed. Thereafter, the user system 112 transmits the task history table 108b to the instruction manual management server 103 via the communication units 107 and 106 (S210), and the user system processing ends.

[0069] After receiving the work history table 108b, the analysis function 105b of the instruction manual management server 103 records the information recorded in the received table in the work history table 104c.

[0070] The terminal 113 displays the instruction manual data in units of work blocks on one screen of the display simultaneously or by scrolling the screen, making it easier for the user to work. At this time, information on the timing of switching the display screen when the user operates the next button 304 can be used as information in the work history table 104c.

[0071] (2-2. Example of operation of matching function) Figure 7 is a flowchart of the processing of the matching function 105a. When the instruction manual management server 103 receives a request for an instruction manual from the user system 112, the processing shown in Figure 7 starts (S601). The matching function 105a first searches the user table 104d for a user who matches the user ID of the requesting terminal, and obtains their characteristic values ​​as a list (S602).

[0072] Then, the instruction manual ID is searched for in the match table 104f using the product ID and the element of the characteristic value list included in the instruction manual request as keys (S603). As an example of the match table 104f, consider a table in which the product ID and characteristic value are used as key columns and the instruction manual ID and priority are used as key values. Here, the priority column is used to uniquely search for the instruction manual with the highest priority value when there are separate instructions for the same product for each of multiple characteristic values.

[0073] For example, consider a match table in which, for the instruction manual for product ID "1," "instruction manual ID = 2, priority = 10" is registered for age = elderly, "instruction manual ID = 3, priority = 5" for gender = female, and "instruction manual ID = 1" for other characteristics.

[0074] In this case, if you search using product ID = 1 and characteristic values ​​= [female, elderly, Japanese], the instruction manual with ID = 2, which has the highest priority, will be found among the instruction manual IDs = 2 and 3, which match the key "product ID = 1, characteristic values ​​= female, elderly."

[0075] Thereafter, the instruction DB 104a is searched for an instruction that matches the instruction ID searched in S603 (S604), and the searched instruction is returned to the requesting terminal (S605), completing the processing of the matching function 105a.

[0076] (2-3. Operation of the Analysis Function) Fig. 8 is a flowchart of the analysis function 105b. The analysis function 105b will be described with reference to the flowchart of Fig. 8. First, the instruction creator 101 starts up the PC 102 and displays an instruction evaluation screen.

[0077] 9 is an image diagram showing an example of an instruction evaluation screen displayed on the PC 102. The instruction evaluation screen includes an instruction ID input field 401 and a search icon 402. The instruction creator 101 inputs the ID of the instruction to be evaluated in the instruction ID input field 401, and then presses the search icon 402 to request an evaluation of the instruction from the instruction management server 103. When the instruction management server 103 receives the evaluation request, processing of the analysis function 105b shown in FIG. 8 is started (S501).

[0078] After starting S501, the analysis function 105b first searches the work history table 104c and the BEM table 104b linked to the instruction manual ID for which the request was received from the storage unit 104 (S502). Then, the analysis function 105b calculates the statistical value of the work history for each work block recorded in the work history table 104c for each characteristic value (S503).

[0079] Next, the statistical values ​​calculated in S503 are compared with the reference values ​​stored in the BEM table 104b for each work block to determine whether the explanation of each work block was appropriate for each characteristic value (S504).

[0080] Next, in S504, the operations contained in multiple work blocks that have been determined to have inappropriate explanations for a user with a certain characteristic value are searched from the BEM table 104b, and operations that are commonly contained in each work block are extracted and determined to be operations with inappropriate explanations (S505).

[0081] Next, the BEM table 104b is searched for operations included in the task block determined in S503 to have inappropriate explanations for the user of a certain characteristic value, and it is determined whether or not those operations include the operation determined in S505 to have inappropriate explanations for the user of a certain characteristic value. If such an operation is found, the task element in that task block that includes that operation is determined to have inappropriate explanations (S506).

[0082] Next, a notification is created to notify the instruction manual creator 101 of the task element for which the explanation is inappropriate, as identified in S506, and is stored in the notification table 104e in association with the instruction manual ID (S507).

[0083] Finally, the instruction manual management server 103 transmits the notification table 104e to the PC 102, and the processing of the analysis function 105b ends.

[0084] (2-4. Specific Operation Example of Analysis Function) The processing of the analysis function 105b will be specifically described below using the assembly of a chest (product ID=1) as an example.

[0085] Assembling a chest is assumed to consist of four work blocks ("Assemble the drawers," "Assemble the storage section," "Attach the top panel," and "Install the casters"), and each work block is broken down into one or more work elements (e.g., "Bolt plate A and plate B together") defined by a combination of an operation and an operation target. Furthermore, there are four types of operations required to assemble a chest ("Screw tightening," "Bonding," "Cam lock fixing," and "Bolt fixing"). An example of this type was previously shown in Figure 2.

[0086] 4 shows an example of the user table 104d, in which the user IDs of five users are associated with their respective characteristic values. In this example, the characteristics for classifying users are gender, age group, and native language, and the possible characteristic values ​​for each are male / female, elderly / middle-aged / young, and Japanese / non-Japanese.

[0087] 3 is an example of the work history table 104c, in which the time (seconds) required to complete each work block is recorded as work history for each user ID. As mentioned above, the time required to complete a work block can be measured as the time from when the explanation for one work block is displayed until the user 114 instructs the display of the next work block, and this timing can be obtained when the user 114 taps the Next button 304 (FIG. 6) on the work block screen of the terminal 113. Note that, for simplicity's sake, five users are analyzed in this example; however, in practice, to obtain more reliable statistics, it is desirable to evaluate the instructions after obtaining the work histories of more users.

[0088] The instruction manual creator 101 starts the PC 102, enters "1" in the instruction manual ID input field 401 on the instruction manual evaluation screen (FIG. 9), and clicks the search icon 402 to start processing of the analysis function 105b (S501). The instruction manual management server 103 searches the work history table 104c and the BEM table 104b in the storage unit 104 for data with an instruction manual ID matching "1" and stores the data in a temporary storage area (S502). The temporary storage area can be the RAM of the control unit 105 or the storage area of ​​the storage unit 104. Data such as T (task block, characteristic value), B (task block, characteristic value), M (operation, characteristic value), and M (operation, characteristic value), which will be described below with reference to FIGS. 11, 13, 15, and 17, are also stored in the temporary storage area. After S502, processing S503 begins.

[0089] FIG. 10 is a flowchart showing an example of the processing in S503, which is processing for determining the average time T (task block, characteristic value) required for a user of each characteristic value to complete each task block.

[0090] Here, matrix X(P, Q) is an NxM matrix with subscripts representing the set of elements in the set P = {p1, p2, ..., pN} and Q = {q1, q2, ..., qM}, and the component in row i and column j is denoted as X(pi, qj). For example, for task block = {"assembly of drawers", "assembly of storage compartments"} and characteristic values ​​= {"male", "female"}, T("task block", "characteristic value") is a 2x2 matrix, and its component T("assembly of drawers", "male") is the average time it took male users to complete assembling a drawer.

[0091] First, the first row of the task history table 104c is extracted, and the user ID and task block completion time list are obtained (S702). Next, the user table 104d is searched using the user ID as a key, and a characteristic value list of the user with the matching user ID is obtained (S703). Next, for each element of the characteristic value list, 1 is added to N (characteristic value), which represents the number of users who have a certain characteristic value (S704).

[0092] Next, the elements of the task block completion time list are added to V(task block, characteristic value), which represents the total time taken by a user with a certain characteristic value to complete a certain task block (S706). After S706, the process returns to S702 and repeats until extraction of all rows in the task history table is complete. Then, T(task block, characteristic value) is obtained by dividing V(task block, characteristic value) by N(characteristic value).

[0093] 11 shows T(task block, characteristic value) calculated based on the user table shown in FIG. 4 and the task history table shown in FIG. 3. For example, there are two users with IDs 1 and 2 who have the characteristic value "elderly" (N(elderly) = 2), and the total time required to complete the task block "drawer assembly" is 1800 (V(drawer assembly, elderly) = 1800), so T(drawer assembly, elderly) = 900. Based on T(task block, characteristic value), the average time T required for users with each characteristic value (e.g., elderly) to complete each task block (e.g., drawer assembly) is determined. After S503 is completed, the analysis function 105b begins processing S504.

[0094] Figure 12 is a flowchart showing an example of the processing in S504, which is a process to calculate a matrix B (task block, characteristic value) that is 1 if the average time taken by a user with a certain characteristic value to complete a certain task block is longer than the expected value, and 0 otherwise.

[0095] First, all elements of B are set to 0 (S802). Next, the first task block row recorded in the BEM table 104b is searched for, and the value of the reference time column is recorded in TM. Next, T (task block, characteristic value) is compared with TM for each characteristic value (S806), and if T (task block, characteristic value) is greater than TM, 1 is recorded in B (task block, characteristic value) (S807). After S807, the process returns to S803 and repeats until all task blocks recorded in the BEM table have been searched.

[0096] FIG. 13 shows an example of B (task block, characteristic value) calculated based on the BEM table 104b shown in FIG. 2 and T (task block, characteristic value) shown in FIG. 11. For example, for the task block "drawer assembly," if the characteristic value is "younger," T (drawer assembly, younger) = 100 < TM = 800, so B (drawer assembly, younger) = 0. Meanwhile, if the characteristic value is "elderly," T (drawer assembly, older) = 900 > 800, so B (drawer assembly, older) = 1. B (task block, characteristic value) identifies task blocks with low task efficiency (i.e., task times longer than the reference time) for users with specific characteristic values. After S504 is completed, the analysis function 105b begins processing S505.

[0097] FIG. 14 is a flowchart showing an example of the processing in S505, which is a process for calculating a matrix M (operations, characteristic values) that represents the proportion of the number of operations included in work blocks that took longer than the reference time out of the number of operations performed by a user 114 of each characteristic value in the entire work for a specified operation.

[0098] First, all elements of M are set to 0 (S902). Next, the BEM table 104b is searched for task blocks that include a certain operation in their operation sequence (S903). Next, for each task block found, the number of operations in the operation sequence is recorded in K (S905), and K is added to L (operation), which represents the number of times a certain operation is performed throughout the entire task (S906). Next, for each characteristic value, it is determined whether B (task block, characteristic value) = 1 (S908). If so, K is added to U (operation, characteristic value), which represents the number of times a certain operation was included in a task block that took longer than the reference time throughout the entire task of the user with a certain characteristic value (S909). After S909, the process returns to S903 and repeats until all operations have been processed.

[0099] FIG. 15 shows an example of M (operation, characteristic value) calculated based on the BEM table 104b shown in FIG. 2 and B (task block, characteristic value) shown in FIG. 13. For example, according to the BEM table 104b, the operation "bolt fastening" is performed once each in the task blocks "drawer assembly," "top panel installation," and "caster installation," resulting in L (bolt fastening) = 3. For the characteristic value "elderly worker," the task blocks for which B (task block, elderly worker) = 1 are "drawer assembly" and "top panel installation," resulting in U (bolt fastening, elderly worker) = 2. Therefore, M (bolt fastening, elderly worker) = U (bolt fastening, elderly worker) / L (bolt fastening) = 0.67 (rounded to the nearest tenth). In the example shown in FIG. 2, the operation "bolt fastening" appears repeatedly throughout all processes. However, the same processing is applied to an operation that appears only once. In this case, M takes two values: "1" or "0."

[0100] In this way, M(operation, characteristic value) is obtained by examining the specific types of operations included in the task that are concentrated in task blocks that have low task efficiency for a specific user. M(operation, characteristic value) identifies operations that may have low task efficiency for a user with a specific characteristic value. After S505 is completed, the analysis function 105b starts processing in S506.

[0101] As described above, in this embodiment, it is possible to identify operations that the user cannot perform appropriately based on information that can be collected by the user 114 simply touching the Next button 304 in Fig. 6. Therefore, it is desirable to revise the instructions by focusing on the operations in question.

[0102] Figure 16 is a flowchart showing an example of the processing in S506, which is a process for calculating a matrix E (work element ID, characteristic value) that is 1 if the work elements that make up a work block where B (work block, characteristic value) = 1 for a user with a certain characteristic value include an operation where M (operation, characteristic value) is greater than a reference value, and 0 otherwise.

[0103] First, all elements of E are set to 0 (S1002). Next, the first task block row is extracted from the BEM table (S1003), and for each characteristic value, it is determined whether B(task block, characteristic value) = 1 (S1005). If so, for each operation of the task element included in the task block (S1006), it is determined whether M(operation, characteristic value) > 0.6 (S1008). If so, E(task element ID, characteristic value) = 1 is recorded (S1009). Note that the reference value of 0.6 can be set arbitrarily. Here, 0.6 is used as the reference value for the majority. After S1009, the process returns to S1003 and is repeated until processing is completed for all task blocks.

[0104] Fig. 17 shows an example of E (task element, characteristic value) calculated based on the BEM table shown in Fig. 2, B (task block, characteristic value) shown in Fig. 13, and M (operation, characteristic value) shown in Fig. 15. In the case of the task block "drawer assembly" and the characteristic value "elderly person", B (drawer assembly, elderly person) = 1, so S1006 is processed.

[0105] The task block "drawer assembly" is composed of task element IDs 1, 2, and 3, which correspond to the operations "bonding," "screw tightening," and "bolt tightening," respectively. Of these, the operation for which M (operation, elderly person) > 0.6 is "bolt tightening," and the task element ID corresponding to the operation "bolt tightening" is "3," so E (3, elderly person) = 1. E (task element, characteristic value) identifies which task element may be the cause of the decline in task efficiency in a task block where a user with a specific characteristic value had low task efficiency. After S506 is completed, the analysis function 105b begins processing S507.

[0106] Figure 18 is a flowchart showing an example of the processing in S507, which is a process in which, for each characteristic value, for each work element that constitutes a work block where B (work block, characteristic value) = 1, if E (work element ID, characteristic value) = 1, a notification is issued informing the user that the description of the work element ID is inappropriate for that characteristic value.

[0107] For example, consider a process for issuing a notification regarding the characteristic value "elderly" based on B (task block, characteristic value) shown in Figure 13 and E (task element ID, characteristic value) shown in Figure 17. The task blocks for which B (task block, elderly person) = 1 are "drawer assembly" and "top panel installation." Of the task elements (IDs = 1, 2, 3) that make up the task block "drawer assembly," the task element ID for which E (task element ID, elderly person) = 1 is "3." Therefore, in S1107, a notification stating "The explanation for task element ID 3 appears inappropriate for elderly people" is issued and saved in the notification table (S1108).

[0108] Similarly, of the task elements that make up the task block "Installing the tabletop," the task element with E (task element ID, elderly person) = 1 is task element ID = 7, so in S1107, a notification message stating "The explanation for task element ID 7 seems inappropriate for elderly people" is issued and saved in the notification message table (S1108). After completing S507, the analysis function 105b performs the process of S508 and returns the notification message table to the requesting PC 102.

[0109] When the evaluation result is received from the instruction management server 103, the PC 102 displays the evaluation result on the instruction evaluation screen. Fig. 9 shows an example of the instruction evaluation screen, in which a notification message 403 is displayed.

[0110] The instruction creator 101 may create multiple instructions for the same product for each characteristic value depending on the content of the notification. For example, if the instruction creator infers from the notification 403 that "it is difficult for elderly people to bolt board A to other boards," the instruction creator may add an image of "the appropriate posture when bolting board A" to the images of task elements 3 and 7 in the instructions for elderly people. In this case, a unique ID (e.g., "2") is assigned to the newly created instruction, a row is added to the match table 104f with product ID = 1 and characteristic value = "elderly people" as a composite primary key, and 2 is recorded in the instruction ID column.

[0111] 9, the results of statistical processing performed on the work history in S503 may be displayed as a graph 409. In this case, the calculated statistical values ​​are stored in the storage unit 104 in S503, and the statistical values ​​are transmitted to the PC terminal 102 in S508.

[0112] As an example, graph 409 displays the value of T (task block, elderly person) together with the reference time as a bar graph for each task block. In this case, a characteristic value selection field 404 may be provided so that the value of T can be displayed for each characteristic value.

[0113] In S503, a process of calculating various statistics (variance, standard deviation, etc.) related to the work history of each work block may be added. In this case, statistics other than the average value may be used in S504, S505, and S506 to identify work elements for which the explanation is inappropriate, or a statistics selection field 406 may be provided so that graphs related to any statistics can be displayed.

[0114] The task history targeted by the analysis function 105b is not limited to task completion time, but may be anything that can be acquired by the user operating the terminal (e.g., the number of times the image 303 is tapped to view, the number of chat questions, etc.). In S503, statistical processing of the various task histories may be performed. In this case, statistical values ​​related to the various task histories may be used in S504, S505, and S506 to identify task elements for which explanations are inappropriate. Furthermore, if the task history is numerical data, a task history selection field 405 may be provided to display statistics related to any task history. Furthermore, if the task history is natural language data such as questions, frequently occurring words may be extracted in S503, and a word cloud 410 may be displayed.

[0115] The content of the notification may include not only the task elements but also task blocks or operations for which the explanations were determined to be inappropriate in steps S505 and S506.

[0116] Identifying task elements with inappropriate instructions may include information not only about the operation but also about the "operation target" commonly used in each task block. For example, if many of the average completion times for task blocks that include board A as an operation target exceed the reference time, task elements that include board A as an operation target may be hindering the appropriate progress of the task. By notifying the user of such task elements, for example, the instruction manual creator may realize that "the user is having difficulty operating board A, the heaviest of the components," and add a note such as "This task requires two people" to the description of the task element that operates component A.

[0117] As described in detail in the above embodiments, the entire assembly process for a product involving assembly is broken down into tasks (task blocks) for assembling a single component of the product and tasks (task elements) each consisting of a single operation and an operation target. The user 114 acquires a task history for each task block from the operations performed by the user 114 on the terminal 113 to view the instruction manual, and stores the acquired task history in the instruction management server 103. The instruction management server 103 determines task blocks with unfavorable trends in the task histories acquired from multiple users as "weak task blocks," determines operations commonly included in multiple "weak task blocks" as "weak operations," and determines task elements that include "weak operations" among the "weak task blocks" as "weak task elements." The server then notifies the user of the determined "weak task elements" as targets for improvement. This allows the user to simply input the reference history for each task block, enabling the instruction creator to determine whether or not the instruction manual needs to be improved for each task element, facilitating the evaluation of the appropriateness of the instruction manual.

[0118] In the above example, furniture assembly instructions were used as an example, but the present invention can be widely applied to any content that explains specific operating procedures to the user, such as cooking instructions or operating instructions for office equipment.

[0119] In the case of instructions for cooking procedures, the instruction DB 104a is composed of a collection of multiple series of instructions for work blocks (e.g., making a roux) that include one or more work elements (e.g., frying flour and butter) that include an operation (e.g., frying) and a component to be operated (e.g., flour and butter) to explain the work procedure for cooking the dish to be cooked (e.g., curry).In the case of an instruction manual for office equipment, the instruction DB 104a is composed of a collection of multiple series of instructions for work blocks (e.g., setup) that include one or more work elements (e.g., downloading a printer driver) that include an operation (e.g., downloading) and a component to be operated (e.g., a printer driver) to explain the work procedure for setting up the office equipment to be cooked (e.g., a printer).

[0120] In this way, it is possible to provide a technology for effectively evaluating manuals that general users refer to. According to the above embodiment, efficient work can be achieved, which reduces energy consumption and carbon emissions, prevents global warming, and contributes to the realization of a sustainable society.

[0121] 103...Instruction management server, 104...Storage unit, 105...Control unit, 112...User system, 113...Terminal

Claims

1. An information processing device having a control unit, a memory unit, an input unit, and an output unit, the control unit having an analysis function, wherein the control unit has a function of providing instructions instructing a user to perform a specified task from the output unit to a terminal operated by the user, the memory unit stores a process metadata table that associates the contents of the instructions with work blocks, work elements associated with the work blocks, and operations that the user should perform with the work elements, and correlates the work blocks, the work elements, and the operations with each other, the input unit acquires a history of processing detected by the terminal when the user uses the instructions for each work block, and records it in the memory unit as a work history table, and the analysis function refers to the process metadata table and the work history table, and analyzes which of the work elements is causing a decrease in work efficiency in a work block where the user has low work efficiency.

2. The information processing device of claim 1, wherein the memory unit has a user table storing characteristic values ​​of the user corresponding to a user ID identifying the user, the process metadata table stores a standard value for evaluating work efficiency for each of the work blocks, and the work history table stores, for a plurality of users, numerical values ​​reflecting the processing history corresponding to a combination of the user ID and the work block, and the analysis function refers to the user table, the process metadata table, and the work history table, aggregates the numerical values ​​in the work history table for each of the users having specific characteristic values, and compares the aggregated result with the standard value, thereby identifying a work block having low work efficiency for the user having the specific characteristic value as one or more work blocks of interest.

3. The information processing device of claim 2, wherein the analysis function refers to the process metadata table and calculates the proportion of each of multiple types of operations included in the instruction manual that are included in the target work block, thereby identifying operations that are unevenly distributed in the target work block as one or more target operations.

4. The information processing device according to claim 3, wherein the analysis function refers to the process metadata table and identifies, among the target work blocks, those to which at least one of the target operations corresponds as the work element, and the output unit outputs information indicating the identified work element.

5. The information processing device according to claim 4, wherein the output section outputs information indicating characteristic values ​​of users with low task efficiency for the target task block together with the identified task elements.

6. The information processing device of claim 1, wherein the input unit acquires the processing history for each work block by detecting, for each work block, at least one of the timing and number of terminal operations performed by the user on the terminal when the user uses the distributed instructions.

7. The information processing device according to claim 6, wherein the processing history detected by said terminal is the timing of display switching when said instruction manual is displayed on said terminal in units of said work blocks.

8. An instruction manual evaluation system including the information processing device of claim 1 and a terminal operated by the user, wherein the terminal displays the instruction manual data in units of work blocks on one screen of the display, either simultaneously or by scrolling the screen, and the processing history detected by the terminal is information on the timing of screen switching on the display.

9. An instruction manual evaluation system including the information processing device of claim 1 and a terminal operated by the user, wherein the processing history detected by the terminal is at least one of information on the number of operations on an icon or image displayed on the display of the terminal and history information of a chat function.

10. An instruction evaluation method executed by an information processing device having a control unit, a memory unit, an input unit, and an output unit, the control unit having an analysis function, the memory unit storing a process metadata table in which the contents of an instruction instructing a user to perform a specific task are associated with a task block, a task element associated with the task block, and an operation to be performed by the user with the task element, and the task block, the task element, and the operation are stored in association with each other; a first step in which the control unit provides the instruction from the output unit to a terminal operated by the user; a second step in which the input unit acquires, for each task block, a history of processing detected by the terminal when the user uses the instruction; a third step in which the memory unit records, for a plurality of users, a work history table storing numerical values ​​reflecting the processing history in association with a combination of the user ID and the task block; a fourth step in which the analysis function refers to the work history table, calculates an average value of the numerical values ​​for each task block, and compares the average value with a reference value to extract one or more task blocks of interest from the task blocks; A method for evaluating instructions which executes the following steps: a fifth step in which the analysis function refers to the process metadata table and identifies, among multiple types of operations included in the instructions, operations that are unevenly distributed in the target work block as one or more target operations; a sixth step in which the analysis function refers to the process metadata table and identifies, among the target work blocks, those to which at least one of the target operations is associated as the work element; and a seventh step in which the output unit outputs information indicating the identified work element.

Citation Information

Patent Citations

  • Implementation method of building block assembling system based on Internet

    CN115826833A

  • Manual management device, manual management system, and control program

    JP2020004274A