Information processing system, information processing method, and information processing program

Abstract

To detect a worker's labor risk with high accuracy using subjective mental and physical information and perform presentation with high reliability and consent feeling for the worker and a manager.SOLUTION: An information processing system includes a processor and a storage device. The storage device holds: subjective mental and physical information of a first user; a detection rule for detecting a labor risk related to a mental and physical condition of the first user; and risk expression information in which content to be presented to the first user and / or a second user is set on the basis of the detected labor risk. The processor extracts subjective mental and physical information in a predetermined period on the basis of the detection rule, detects a labor risk on the basis of the detection rule and a statistic of the subjective mental and physical information in the predetermined period, generates the content to be presented to the first user and / or the second user related to the detected labor risk on the basis of the risk expression information, and outputs the generated presentation content to the first user and / or the second user to which the content is to be presented.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technology for detecting risks during work from subjective mind-body related information and supporting the management of employees, etc. [Background technology]

[0002] In recent years, worker health management has become an important issue, and both objective assessments using biometrics and subjective assessments using interviews are attracting attention. For example, health management for workers whose work is demanding and potentially dangerous, such as drivers, is considered more important than for other occupations. In assessing subjective physical and mental information, in addition to interviews conducted during regular health checkups, there are technologies that acquire and record users' subjective physical and mental information from smartphone applications. Answers to physical and mental information such as subjective stress and fatigue can be measured using indicators such as the Visual Analogue Scale (VAS), which can intuitively express the degree of fatigue and stress felt by individuals.

[0003] Japanese Patent Application Laid-Open No. 11-169362 (Patent Document 1) states that "the subject is encouraged to clearly become aware of the agreement or discrepancy between the subjective assessment of their mental and physical state and the objective assessment of their mental and physical state through measurement." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 11-169362 Summary of the Invention [Problem to be solved by the invention]

[0005] The information (mental and physical state judgment value) dealt with in the above patent documents is primarily based on objective mental and physical information (objective mental and physical state judgment value), or subjective mental and physical information (subjective mental and physical state judgment value) is compared with objective mental and physical information (objective mental and physical state judgment value) to achieve highly accurate risk detection for subjects.

[0006] However, subjective physical and mental information varies greatly from person to person, and the severity of the information and numerical values ​​provided for the individual and their actual working situation varies greatly from person to person. Furthermore, if an individual wants to lie, they can enter false information, raising issues with reliability. Therefore, the above-mentioned literature has not attempted to resolve the issues specific to subjective physical and mental information. Furthermore, when displaying an alert to a subject based on risk detection results, the information and format of presentation required may differ depending on the position of the person receiving the alert (e.g., worker versus manager), but no particular consideration has been given to such differences.

[0007] Therefore, an object of the present invention is to detect labor risks using subjective physical and mental information, and when presenting a preset alert for the detected labor risks, to present alert content that is effective and easy for workers and / or their managers to comply with.More preferably, an object of the present invention is to collect evaluations of the alert from at least one or both of the worker and their manager, and update the labor risk and alert settings based on the evaluations, thereby improving the reliability of labor risk detection and the sense of acceptability of the alert content. [Means for solving the problem]

[0008] In order to solve at least one of the above problems, the present invention provides an information processing system comprising a processor and a storage device, wherein the storage device stores past subjective physical and mental information input by a first user regarding his or her own physical and mental state at any point in time, detection rules for detecting labor risks related to the first user's physical and mental state based on the subjective physical and mental information, and risk expression information that sets content to be presented to the first user and / or content to be presented to a second user different from the first user based on the detected labor risk, and the processor executes the following steps: a first step of extracting the subjective physical and mental information for a predetermined period based on the detection rules; a second step of detecting the labor risk based on statistics of the subjective physical and mental information for the predetermined period and the detection rules; a third step of generating content to be presented to the first user and / or content to be presented to the second user regarding the detected labor risk based on the risk expression information; and a fourth step of outputting the generated content to the first user and / or the second user to whom the content is to be presented. [Effects of the Invention]

[0009] Therefore, according to one aspect of the present invention, worker labor risks can be detected with high accuracy using subjective physical and mental information, and highly reliable and convincing information can be presented to both the worker and the manager.

[0010] The details of at least one implementation of the subject matter disclosed herein are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the disclosed subject matter will become apparent from the following disclosure, drawings, and claims. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram illustrating an example of the configuration of a subjective mind-body information processing system according to an embodiment of the present invention. [Figure 2] 1 is a flowchart illustrating an example of the present invention, showing an outline of processing performed in the subjective mind-body information processing system. [Figure 3] FIG. 2 is a diagram illustrating an example of subjective mind-body information according to the embodiment of the present invention. [Figure 4] FIG. 10 illustrates an example of biometric information according to the embodiment of the present invention. [Figure 5] FIG. 10 illustrates an example of labor information according to the embodiment of the present invention. [Figure 6] FIG. 10 illustrates an example of incident information according to the embodiment of the present invention. [Figure 7] FIG. 10 is a diagram illustrating an example of business performance information according to the embodiment of the present invention. [Figure 8] 5 is a flowchart illustrating an example of processing performed by the data processing / storage unit according to the embodiment of the present invention. [Figure 9] 10 is a graph illustrating an example of heart rate data according to an embodiment of the present invention. [Figure 10] 1 is a graph illustrating an example of heart rate fluctuation according to an embodiment of the present invention. [Figure 11] 10 is a graph illustrating an example of the spectral power density of heart rate variability according to an embodiment of the present invention. [Figure 12] 10 is a flowchart illustrating an example of processing performed by the time-series data extraction unit according to the embodiment of the present invention. [Figure 13] 10 is a flowchart illustrating an example of processing performed by the risk detection unit according to the embodiment of the present invention. [Figure 14] FIG. 10 is a diagram illustrating an example of detection rule setting information according to the embodiment of the present invention. [Figure 15] FIG. 2 is a diagram illustrating an example of labor risk information according to an embodiment of the present invention. [Figure 16] 10 is a flowchart illustrating an example of processing performed by a presentation content setting unit according to the embodiment of the present invention. [Figure 17] FIG. 10 is a diagram illustrating an example of risk expression information according to the embodiment of the present invention. [Figure 18] 10 is a graph illustrating an example of labor risk detection result information presented by the system according to an embodiment of the present invention. [Figure 19]10 is a flowchart illustrating an example of processing performed by the detection and presentation effect acquisition unit according to the embodiment of the present invention. [Figure 20] FIG. 2 illustrates an example of an information collection medium used to obtain a detection and presentation effect according to an embodiment of the present invention. [Figure 21] FIG. 3 is a diagram illustrating an example of detection and presentation effect information according to the embodiment of the present invention. [Figure 22] 10 is a flowchart illustrating an example of processing performed by the detection rule update unit according to the embodiment of the present invention. [Figure 23] 10 is a flowchart illustrating an example of processing performed by the risk expression update unit according to the embodiment of this invention. [Figure 24] FIG. 10 is a diagram illustrating an example in which risk expression information is updated according to the embodiment of the present invention. [Figure 25] FIG. 10 illustrates an embodiment of the present invention, showing an example of an alert for a worker presented by a subjective mind-body information processing system, and an example of an information collection medium for acquiring the detection and presentation effect. [Figure 26] FIG. 10 illustrates an example of an updated alert for a worker that is presented by the subjective mind-body information processing system after the risk expression information is updated according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

[0013] <System configuration> 1 is a block diagram showing an example of the configuration of a subjective physical and mental information processing system according to an embodiment of the present invention. The subjective physical and mental information processing system of this embodiment includes an in-work labor support server 1. The in-work labor support server 1 collects, via a network 12, information related to a worker's work, subjective physical and mental information that is information related to the worker's subjective physical and mental state, biometric information about the worker, and work information about the worker, estimates risks related to the worker's safety and physical and mental health (hereinafter referred to as labor risks), and notifies the worker if the estimated labor risk exceeds a threshold value.

[0014] Note that a "worker" is a person under management and is an example of a user category of the subjective physical and mental information processing system in this embodiment. As described above, the worker's labor risk is estimated based on the subjective physical and mental information input by the worker and biometric information obtained by measuring the worker, and the result is presented to the worker himself / herself. An example of a worker is a truck driver in a logistics business, but it may also be someone engaged in any other work, such as a loading and unloading worker or a factory worker.

[0015] Furthermore, the users of the subjective mind-body information processing system in this embodiment may include users in a category other than the above-mentioned "worker." One example is a "manager." A manager is a person in a position to manage workers, and may be, for example, a supervisor with the authority to give instructions and commands to workers, a person who manages the safety of the workers' work, or an industrial physician who is involved in the health management of workers. Users may be classified into three or more categories, such as truck drivers, their supervisors, and industrial physicians. Alternatively, the workers may be students, and the manager may be a teacher.

[0016] An alert (presentation content) based on the estimated labor risk for a worker is presented to the worker himself / herself as well as to the manager. As will be described later, the alert presented to the manager may contain information different from that contained in the alert presented to the worker himself / herself, and the method of presenting the alert may be different from that for the worker himself / herself. This is because the responsibilities, authority, and information held as a premise may differ depending on the user's position, and the information and presentation format required may differ accordingly.

[0017] The network 12 is connected to a work-related information server 8 that stores work-related information about workers, a subjective physical and mental information collection device 9 that acquires subjective physical and mental information about workers, a biometric information collection device 10 that acquires biometric information about workers, and a prediction result display / input terminal 11 that outputs notifications from the in-work labor support server 1, and is capable of communicating with the in-work labor support server 1.

[0018] The subjective mental and physical information collection device 9 is a device that collects the subjective mental and physical information of the worker. The subjective mental and physical information collection device 9 can be a smartphone, a tablet, or any other device that has an operable LCD panel or physical buttons.

[0019] The subjective mental and physical information collecting device 9 can collect information about the worker's state of mind through a questionnaire or interview format. For example, the subjective mental and physical information collecting device may ask the worker about the stress, fatigue, drowsiness, anxiety, and other indicators that may affect the worker's work, and the worker can use the subjective mental and physical information input unit 91 to respond to the degree of stress or other conditions using a binary "yes / no" or several levels. Alternatively, a slide bar may be displayed on the subjective mental and physical information collecting device, allowing the worker to respond with a level ranging from 0 to 100, like a visual analogue scale (VAS).

[0020] The biometric information collection device 10 includes a heart rate monitor 101 that detects heart rate data, a thermometer 102 that detects the worker's body temperature, a sphygmomanometer 103 sensor that detects the worker's blood pressure, and a pulse oximeter 104 that detects arterial oxygen saturation (SpO2). In addition to a stationary device, the biometric information collection device 10 can be a wearable device that can be worn by a worker, a sensing device installed at the workplace such as a chair used during work, or an image recognition system that captures and analyzes the facial expressions or behavior of a worker.

[0021] The sensors of the biometric information collecting device 10 are not limited to those described above, and sensors that detect sweat rate, body temperature, blinking, eye movement, brain waves, etc. may be used. The biometric information collecting device 10 can also set an identifier that identifies the worker and add the identifier to various sensing data.

[0022] The work-related information server 8 includes work performance information 81, environment information 82, behavior information 83, attendance information 84, and incident information 85 of the worker.

[0023] The work performance information 81 is information about the work performance of a worker. The work performance information 81 may also include information about the achievement status of work quotas and deliveries. The work performance information 81 may also include specific information according to a specific industry or occupation. For example, if the occupation is a delivery driver, the information may include whether or not there have been any delivery delays.

[0024] The environmental information 82 is information about the environment that may affect the work of the worker. For example, the environmental information 82 includes the climate and temperature at the work site. In addition, specific information may be included depending on a specific industry or occupation. For example, if the occupation is a delivery driver, the environmental information 82 may include the traffic congestion situation on the delivery route.

[0025] The behavior information 83 is information about the behavior of the worker during work, and includes the details of the work performed each day written by the worker in a daily report, diary, etc. Furthermore, the physical behavior log information of the worker may be obtained from acceleration and position information measured by a wearable device that can be worn by the worker. Furthermore, the specific work content estimated from the behavior log information may be used as the behavior information 83.

[0026] The attendance information 84 is information about the attendance of the worker, and includes whether or not the worker has attended work each day. The attendance information 84 can also include whether or not the worker has been late, absent due to illness, or absent without permission.

[0027] The incident information 85 is information related to accidents or industrial accidents that have occurred around the worker during work, and situations that could escalate into such accidents (near misses). It may also include biometric information measured by a wearable device included in the biometric information collection device 10, or abnormal values ​​in the behavior log information included in the behavior information 83. It may also include specific information according to a specific industry or occupation. For example, if the occupation is a transportation driver, a near miss during driving work may be estimated based on sudden changes in speed or acceleration measured by a vehicle behavior sensor mounted on the vehicle, and this may be used as the incident information 85.

[0028] The prediction result display / input terminal 11 includes an output unit 111 and an input unit 112. The output unit 111 outputs labor risk warnings and attention calls sent from the in-work labor support server 1 to the user, i.e., the manager or worker. The input unit 112 accepts text data selected or input by the user himself / herself, indicating how effective the labor risks and their expression methods output by the output unit 111 were for the user, i.e., the manager or worker, and transmits this to the in-work labor support server 1. The input unit 112 can accept voice input in addition to text data.

[0029] The prediction result display / input terminal 11 may be a mobile terminal carried by a worker or manager, a wearable device that can be worn by a worker or manager, or a computer installed in a workplace. The prediction result display / input terminal 11 may also include specific information according to a specific industry or occupation. For example, if the occupation is a transportation driver, the prediction result display / input terminal 11 may be a car navigation device installed in a vehicle.

[0030] The in-office labor support server 1 is a computer including a processor 2, a memory 3, a storage device 4, a communication interface 5, an output device 6, and an input device 7. The memory 3 loads each of the following functional units as programs: a data processing and storage unit 31, a time-series data extraction unit 32, a risk detection unit 33, a presentation content setting unit 34, an alert presentation unit 35, a detection and presentation effect acquisition unit 36, a detection rule setting and update unit 37, and a risk expression update unit 38. Each program is executed by the processor 2. Details of each functional unit will be described later.

[0031] The processor 2 operates as a functional unit that provides a predetermined function by executing processing in accordance with the program of each functional unit. For example, the processor 2 functions as a risk detection unit 33 by executing a risk detection program. The same applies to other programs. Furthermore, the processor 2 also operates as a functional unit that provides each function of multiple processes executed by each program. A computer and a computer system are devices and systems that include these functional units.

[0032] The storage device 4 stores data used by the above-mentioned functional units, including subjective physical and mental information 41, labor risk information 42, detection rule setting information 43, risk expression information 44, detection and presentation effect information 45, and external information 46.

[0033] The external information 46 includes biometric information 461, labor information 462, and incident information 463. and business performance information 464. The labor information 462 includes the remaining data from the various data stored in the business-related information server 8, excluding the incident information 463 and the business performance information 464. Details of each data stored in the storage device 4 will be described later.

[0034] The input device 7 includes a mouse, keyboard, touch panel, etc. The output device 6 includes a display, speaker, etc. The communication interface 5 is connected to a network 12 and communicates with a work-related information server 8, a subjective physical and mental information collection device 9, a biometric information collection device 10, and a prediction result display / input terminal 11.

[0035] <Software configuration> The data processing and storage unit 31 acquires and converts subjective mental and physical information from the subjective mental and physical information collection device 9 and stores it in subjective mental and physical information 41, acquires and converts biometric information from the biometric information collection device 10 and stores it in biometric information 461, acquires incident information 85 and work performance information 81 from the work-related information server 8 and stores them in incident information 463 and work performance information 464, respectively, and acquires and converts environmental information 82, behavioral information 83 and attendance information 84 from the work-related information server 8 and stores them in labor information 462.

[0036] The data processing and storage unit 31 calculates the R wave interval (RRI = RR Interval) of the heart rate data from the data of the heart rate meter 101 (hereinafter referred to as heart rate data) among the biometric information, and calculates an autonomic nervous function (ANF) index from the RRI data (or heart rate interval data) and stores it in the biometric information 461.

[0037] Furthermore, the data processing and storage unit 31 converts the data obtained from the subjective mental and physical information collection device 9 into indicators and the like according to predetermined rules, and stores the converted data in the subjective mental and physical information 41. For example, if the subjective mental and physical information is data asking about emotions such as "anger" or "anxiety," each indicator is converted according to predetermined rules into indicators related to subjective mental and physical states such as "stress" or "fatigue." The rule may be such that a threshold is manually set for each indicator of the subjective mental and physical information, and if the indicator exceeds this threshold, the subjective mental and physical state is deemed to be poor, or such that if a response value that is more extreme than normal is detected through statistical analysis, the subjective mental and physical state is deemed to be poor.

[0038] Furthermore, the data processing and storage unit 31 extracts data related to the work content performed each day from the environmental information 82, behavioral information 83, and attendance information 84 held by the work-related information server 8, and collates each piece of data and stores it in the labor information 462.

[0039] The time-series data extracting unit 32 specifies a specific index from the subjective psychosomatic information 41 and the external information 46 in accordance with the detection rule setting information 43, and extracts data for a specific time interval.

[0040] The risk detection unit 33 detects labor risks within the work from the subjective physical and mental information 41 and external information 46 extracted by the time-series data extraction unit 32 in accordance with the detection rule setting information 43 and stores the detected labor risks in labor risk information 42 .

[0041] The presentation content setting unit 34 determines the presentation content of the detected labor risk information 42 in accordance with the risk expression information 44.

[0042] The alert presentation unit 35 outputs an alert from the prediction result display / input terminal 11 with the presentation content determined by the presentation content setting unit 34 regarding the detected labor risk information 42 .

[0043] The detection and presentation effect acquisition unit 36 ​​acquires from the prediction result display and input terminal 11 how effective the alert output by the prediction result display and input terminal 11 was for the user, that is, the manager or worker, and stores this in detection and presentation effect information 45.

[0044] The detection rule setting update unit 37 analyzes the detection and presentation effect information 45 and updates the existing detection rule to a more reliable detection rule.

[0045] The risk expression update unit 38 analyzes the detection and presentation effect information 45 and updates the existing risk expression to a risk expression that is more convincing to the worker or manager.

[0046] <Processing Overview> 2 is a flowchart showing an outline of the processing performed in the subjective mental and physical information processing system. First, the data processing and storage unit 31 of the in-work labor support server 1 extracts and converts various data from the biometric information collection device 10, the subjective mental and physical information collection device 9, and the work-related information server 8, and stores the data in the storage device 4 of the in-work labor support server 1 (S201).

[0047] The biological information in this embodiment may be, for example, a power spectrum density (described later) calculated from the worker's heart rate data, or an autonomic nervous index (described later) based on the NN interval (the difference between the R-wave intervals) calculated from time domain analysis. Alternatively, results analyzed and calculated from the autonomic nervous index may be used.

[0048] The data processing and storage unit 31 performs preprocessing of the biological information 461 by excluding or interpolating missing sections of the heart rate data (RRI data). For example, if the length of a missing section exceeds a predetermined threshold Th, the data processing and storage unit 31 can exclude the heart rate data of that section, and if the length of the missing section is equal to or less than the predetermined threshold Th, perform interpolation. Next, the data processing and storage unit 31 calculates an autonomic nervous function index (ANF information) from the preprocessed heart rate data as described below, and stores the information in the biological information 461.

[0049] The data processing and storage unit 31 extracts the incident information 85 and the work performance information 81 from the work-related information server 8, and stores them respectively in the incident information 463 and the work performance information 464 of the external information 46. Furthermore, the data processing and storage unit 31 extracts data related to the content of work performed each day from the environmental information 82, behavioral information 83, and attendance information 84 from the work-related information server 8, collates each piece of data, and stores it in the labor information 462.

[0050] The time-series data extracting unit 32 extracts data of a specific index from the subjective physical and mental information 41 and the external information 46 in accordance with an existing detection rule setting for a specific time period (S202).

[0051] Next, the risk detection unit 33 detects labor risks in accordance with existing detection rule settings (S203).

[0052] Next, the presentation content setting unit 34 acquires the presentation method and presentation wording set in the existing risk expression information for the labor risk detected by the risk detection unit 33, and sets the alert content including the labor risk information (S204).

[0053] Next, the alert presentation unit 35 transmits the alert content set by the presentation content setting unit 34 to the prediction result display / input terminal 11 of the relevant worker or their manager (S205).

[0054] When notifying labor risk information, in addition to the content of the labor risk information, the message also includes indicators of the explanatory variables set in the detection rules as the basis for issuing the warning, making it possible to notify an alert that is easy for workers or managers to understand.

[0055] The detection and presentation effect acquisition unit 36 ​​asks the user, i.e., the worker or manager, how convincing the content of the issued alert was for the labor risk information sent to the prediction result display and input terminal 11, sends a message to the prediction result display and input terminal 11 prompting the user to respond, and receives input (detection and presentation effect) from the worker or manager from the input unit 112 of the prediction result display and input terminal 11, and acquires information indicating an evaluation of the presented content, which is the alert (S206).

[0056] The detection and presentation effect acquisition unit 36 ​​allows the worker or manager to input the degree of satisfaction with an alert notified by the in-work labor support server 1, and the degree of satisfaction is stored as the alert effect in the detection and presentation effect information 45. This allows the alerts output by the in-work labor support server 1 to be updated to warnings that are easy for the worker or manager to understand and more reliable. That is, the detection rules are updated (S207), and the detection and presentation effects are analyzed, and the risk expressions are updated to increase the degree of satisfaction (S208). Note that, for example, if the worker is a transport driver, the input from the input unit 112 may be made periodically, such as after the worker finishes driving or after the worker finishes work each day, or at the end of each week, the end of each month, quarterly, or semi-annually.

[0057] 25 is a diagram showing an example in which an alert from the in-work labor support server 1 and a questionnaire asking about the effectiveness of the alert are output on the prediction result display / input terminal. In the example shown, an alert screen 2500 is presented on the prediction result display / input terminal 11, and a message 2501 indicating that the worker's interview response values ​​(subjective physical and mental information) have dropped significantly over the past few days and that there has been a day when the number of near misses has increased sharply, and a graph 2502 visualizing the message 2501 are output.

[0058] By referring to the message 2501 and the graph 2502, the worker or manager can understand the warning and its cause, increasing the possibility that the worker or manager will intuitively follow the issued alert. In addition to the warning message, a suggestion for reducing risk may be added to the message 2501.

[0059] Furthermore, the prediction result display / input terminal 11 presents a questionnaire screen 2510 asking whether the presented alert was effective or not, either simultaneously with the presentation of the alert or at a later timing. The prediction result display / input terminal 11 first presents a question 2511 asking whether the presented alert is acceptable, and collects the answer in an answer field 2512. The answer collected in the answer field 2512 may be a binary choice of Yes / No, may be divided into multiple levels, may be a VAS with a granularity of about 100 levels, or may be in the form of a continuous value. Furthermore, if the worker or manager answers that the alert is acceptable, the prediction result display / input terminal 11 presents a question 2513 asking what factors the worker or manager feels actually caused the labor risk to be detected, and collects the answer in an answer field 2514. The answer may be collected in the answer field 2514 by text input or voice input. In addition to such questions, questions asking for reasons for dissatisfaction and questions asking for external information to be presented along with the alert may also be presented.

[0060] FIG. 26 is a diagram showing an example in which an alert detected by the in-work labor support server 1 is output on a prediction result display / input terminal after the risk expression has been updated by the detection / presentation effect acquisition unit. The alert screen 2600 includes an updated message 2601 and an updated graph 2604. The updated message 2601 includes new explanatory information 2602 added as a result of updating the risk expression. Countermeasure information 2603 may also be included. Similarly, new information 2605 that increases the persuasiveness of the alert is added to the updated graph 2604. Alternatively, as a result of updating the risk expression, the existing information may be rejected as being too complicated. In such cases, the updated message 2601 and the updated graph 2604 will have simpler content.

[0061] The in-work labor support server 1 of this embodiment can eliminate individual differences and present highly reliable labor risk information by analyzing time-series changes in subjective physical and mental information together with external information. Furthermore, by collecting and reanalyzing the alert presentation effects of detected labor risks, it is possible to issue more reliable alerts that are more convincing to workers or managers.

[0062] For workers or managers, it is difficult to understand why an alert was issued when they are suddenly notified that a labor risk has been detected, primarily based on subjective physical and mental information. Furthermore, the reliability of subjective physical and mental information data can be an issue depending on the scale of the person entering the information and the subjectivity of the response. Therefore, the in-work labor support server 1 of this embodiment collects and reanalyzes the effects on the user when an alert is issued, enabling it to issue highly reliable alerts that are easy for workers or managers to understand and accept.

[0063] <Data details> Next, details of the data used by the in-business labor support server 1 will be described.

[0064] 3 is a diagram showing an example of subjective mental and physical information 41 acquired by subjective mental and physical information collecting device 9 and processed and stored in the data processing and storage unit. The subjective mental and physical information 41 includes a user ID 411, date and time 412, and a response to a medical questionnaire in one record. The response to the medical questionnaire is information related to subjective mental and physical information, and may include, for example, a response value 413 to a medical questionnaire asking about fatigue and / or a response value 414 to a medical questionnaire asking about the soundness of the most recent sleep.

[0065] The user ID 411 stores the identifier of the worker. In this embodiment, it is assumed that an identifier set in advance in the subjective mental and physical information collecting device 9 is used. The date and time 412 stores the date and time when the subjective mental and physical information collecting device 9 measured the data.

[0066] 4 is a diagram showing an example of biometric information 461 measured and processed by the biometric information collecting device 10. The biometric information 461 stores a user ID 411, a measurement time 4611 of the biometric information, an average heartbeat interval 4612 measured by the heart rate meter 101, a TP 4613 calculated from the heartbeat interval (RRI data) measured by the heart rate meter 101, an LF / HF 4614, a blood pressure 4615 measured by the sphygmomanometer 103, a body temperature 4616 measured by the thermometer 102, and an arterial blood oxygen saturation (SpO2) 4617 measured by the pulse oximeter 104. Note that "not measured" is stored for items for which data was not measured.

[0067] LF / HF 473, which is one of the autonomic nervous function indices, may be objective mind-body information calculated from biological information 461, and as will be described later, is the ratio of low frequency (LF) components to high frequency (HF) components of the power spectrum density of the R wave interval (RRI) of heart rate data, and is stored as a value indicating the balance of the autonomic nervous system (sympathetic and parasympathetic nerves). Note that the low frequency component indicates an activity index of the sympathetic nerve, and the high frequency component indicates an activity index of the parasympathetic nerve.

[0068] 5 is a diagram showing an example of labor information 462. The labor information 462 includes, in one record, a base ID 4621, a user ID 411, a work date 4622, weather 4623, a busy period 4624, a number of consecutive work days 4625, working hours 4626, and break times 4627.

[0069] The base ID 4621 stores an identifier of the area where the worker who obtained the data works. The working date 4622 stores the working date of the worker. The weather 4623 stores weather information for the area corresponding to the base ID 4621. The busy period 4624 stores busy information corresponding to the base ID 4621 and the working date 4622. For example, the busy period 4624 stores information indicating whether the day indicated by the working date 4622 is included in the busy period (Yes) or not (No) at the base indicated by the base ID 4621. The number of consecutive working days 4625 stores the number of consecutive working days up to the current point, extracted from the attendance information 84 of the work-related information server 8. The working hours 4626 and the break times 4627 store the working hours and break times extracted from the attendance information 84 of the work-related information server 8.

[0070] 6 is a diagram showing an example of incident information 463 stored in the external information 46 of the business-related information server 8. The incident information 463 includes, in one record, a base ID 4621, a user ID 411, a work date 4622, a number of near misses 4634, whether an accident has occurred 4635, and whether an industrial accident has occurred 4636.

[0071] The incident information 463 stores information about incidents that occurred around the worker identified by the user ID 411 during work. The incident information 463 includes information about whether or not an industrial accident occurred during work 4636 and the number of near misses 4634. The incident information 463 may also include biometric information measured by a wearable device included in the biometric information collection device 10 and abnormal values ​​of action log information included in the action information. In addition, the incident information 463 may include specific information depending on a specific industry or occupation. For example, if the occupation is a transportation driver, incidents during driving work may be automatically estimated using a program, machine learning model, etc., based on sudden changes in speed or acceleration measured by a vehicle behavior sensor installed in the vehicle, and the incident information may be used.

[0072] FIG. 7 is a diagram showing an example of business performance information 464 stored in the external information 46 of the business-related information server 8. The business performance information 464 includes, in one record, a base ID 4621, a user ID 411, a work date 4622, and other performance information. The performance information may include specific information according to a specific industry or occupation. For example, if the occupation is a delivery driver, the performance information may include a delivery route 4641 and whether or not there was a delay 4642. The performance information may also include the number of near misses 4634.

[0073] The delivery route is stored in the delivery route 4641. The delay status 4642 stores whether or not a delay was reported at the time of delivery.

[0074] 14 is a diagram showing an example of the detection rule setting information 43. The detection rule setting information 43 includes a rule ID 431, an explanatory variable 432, a target variable 434, a determination formula 435, and a threshold 436. Furthermore, like the second explanatory variable 433, a plurality of explanatory variables may be stored.

[0075] The rule ID 431 stores an identifier of a rule for detecting labor risks. The explanatory variables 432 and 433 store explanatory variables used in various rules. For example, the TP 4613 and LF / HF 4614 stored in the biometric information 461, and the answer values ​​413 and 414 to the medical interview stored in the subjective physical and mental information 41 can be used as explanatory variables. In addition, information related to the work environment such as the number of consecutive days worked 4625 and working hours 4626 from the labor information 462 can also be used. The objective variable 434 stores an index that defines labor risks. For example, the objective variable 434 can be the probability of occurrence of a work error such as not achieving a quota calculated from work performance information (work error risk), the probability of a near miss occurring during work (near miss risk), etc.

[0076] The judgment formula 435 is a relational expression between the explanatory variables 432 and 433 and the objective variable 434. The judgment formula 435 may include a threshold 436. The judgment formula 435 and the threshold 436 can be set manually or automatically by statistical analysis of previously accumulated subjective physical and mental information and external information. Alternatively, the judgment formula 435 may be created by defining risk from fluctuations in the explanatory variables using a mechanically determined threshold without data analysis. Furthermore, the judgment formula 435 and the threshold 436 may be set by a machine learning model.

[0077] The detection rule setting information 43 shown in FIG. 14 is an example, and the in-work labor support server 1 can store any type of rule for detecting labor risks. For example, instead of the judgment formula 435 and threshold value 436 shown in FIG. 14, any model that calculates a target variable based on one or more explanatory variables may be stored. The explanatory variables may include any item of the subjective physical and mental information 41, any item of the external information 46, or a combination of multiple items thereof. Furthermore, statistics of the values ​​of the subjective physical and mental information 41 or the external information 46 can be used as explanatory variables. The statistics are, for example, the average value or variation (e.g., variance, standard deviation), but may also be something else (e.g., the difference between any time points, the difference between the maximum value and the minimum value, etc.). It may also be the difference between any time points, etc.

[0078] Specifically, for example, a rule may be set to detect when a medical interview response value falls below a predetermined reference value within a predetermined period. Alternatively, a rule may be set to detect when the average medical interview response value for a predetermined period is below a predetermined reference value and the average value in the latter half of the period is lower than the average value in the first half of the period. Alternatively, a rule may be set to detect when the standard deviation in the latter half of the period is larger than the standard deviation in the first half of the period and the average value in the latter half is lower than the average value in the first half of the period. In this case, the predetermined period may be set to a relatively long period (e.g., 8 weeks) or a relatively short period (e.g., 4 weeks).

[0079] In this way, by arbitrarily setting the type of statistics to be used as explanatory variables, the length of the data acquisition period for calculating the statistics, the comparison target for the statistics, etc., it is possible to set rules to detect various changes that may be correlated with labor risk, such as sudden short-term changes in values ​​and long-term changes in values.

[0080] 15 is a diagram showing an example of labor risk information 42. The labor risk information 42 accumulates risk information detected by each rule in the detection rule setting information 43. The labor risk information 42 includes an assessment date 421, a user ID 411, a rule ID 431, a data extraction start point 422, a data extraction end point 423, and a risk score 424 in one record.

[0081] The date on which the labor risk was detected is stored in the assessment date 421. The data extraction start point 422 and the data extraction end point 423 store, in date format, the extraction start point and extraction end point of the individual's subjective physical and mental information and external information used for detection, respectively.

[0082] The risk score 424 is a value output as a result of applying a judgment formula to the extracted time-series data of subjective physical and mental information and external information. The risk score 424 stores a value that represents the probability of an accident, incident, work error, or other labor risk occurring, with a minimum value of 0 and a maximum value of 1.

[0083] 17 is a diagram showing an example of the risk expression information 44. The risk expression information 44 stores the content to be presented to a worker or a manager when a risk detected by each rule in the detection rule setting information 43 is presented. The risk expression information 44 includes a rule ID 431, a presentation target 441, presentation information 442, a presentation method 443, and a presentation statement 444 in one record.

[0084] The presentation target user 441 is information indicating the category of the user to be presented, and in this embodiment there are two types: "worker" and "manager." By storing multiple categories in the presentation target user 441, it is possible to separate the presentation of risks according to the user's position and set them in detail.

[0085] The presentation information 442 includes explanatory variables 432 and 433 and a target variable 434 set in each rule of the detection rule setting information 43. In addition, the presentation information 442 may also include other indicators that may improve the effectiveness of an alert when presented, such as weather or the number of consecutive days a worker has worked.

[0086] The presentation method 443 stores information about the medium used to present the alert. The presentation method 443 includes, for example, "text" and "graph." Other methods such as music and light may also be used.

[0087] The presentation message 444 stores a message to be presented to the user when an alert is issued, which is determined for each rule ID 431. The presentation message 444 may include information about each piece of data specified in the presentation information 442.

[0088] FIG. 20 is a diagram showing an example of a user response result 14. In this embodiment, the information collection medium is a questionnaire format. The user response result 14 is information obtained by the detection and presentation effect acquisition unit 36. For example, after an alert is presented to a worker or a manager, the worker or manager is asked to what extent the alert was convincing to them, and the obtained result is stored in the user response result 14. The user response result 14 includes, in one record, an implementation date 141, a respondent ID 142, a presentation target 441, a question ID 143, a question statement 144, an answer 145, and a comment 146 from the respondent.

[0089] The implementation date 141 stores the date on which information collection was carried out. The respondent ID 142 stores the user ID of the person to whom the question is to be asked. The question is asked to the worker or manager who is the user for whom a labor risk has been detected and an alert has been presented.

[0090] The question ID 143 is an identifier of the question. The question 144 stores a statement that cites the wording and graphs of each detection rule defined in the risk expression information 44 and asks whether the alert is convincing.

[0091] The answer 145 is stored as a binary choice of yes or no. Alternatively, it may be classified into levels using integer values ​​of about 1 to 5, or may be in a format in which a numerical value or continuous value of about 0 to 100 obtained by a VAS is stored.

[0092] Comments 146 from the respondent store supplementary information obtained in addition to the answer 145. For example, comments 146 from the respondent include reasons for low satisfaction, information that the respondent wants to confirm other than the explanatory variables and the objective variables, and explanations of situations when workers actually felt a labor risk.

[0093] 21 is a diagram showing an example of detection and presentation effect information 45. The detection and presentation effect information 45 aggregates response results 14 from users and stores how effective each detection rule and its risk expression is for the user. The detection and presentation effect information 45 includes, in one record, a rule ID 431, an information acquisition date 451, presentation target persons 441, the number of target persons 452, the percentage of those who answered "effective" 453, the need for improvement 454, a proposal for adding explanatory variables 455, and a proposal for improving the expression method 456.

[0094] The information acquisition date 451 stores the date on which the answer results 14 from the users were collected and each piece of information in the detection and presentation effect information 45 was acquired.

[0095] The number of subjects 452 stores the number of people in each group when the information on the answer results 14 from users is grouped and tallied by the implementation date 141, the presented subjects 441, and the question ID 143.

[0096] The percentage of people who answered "effective" 453 stores the percentage of people who answered "I felt convinced" in answer 145 of answer result 14 from users out of the total number of subjects 452.

[0097] The necessity for improvement 454 stores "yes" when the ratio of those who answered "yes" 453 is below a predetermined threshold, and stores "no" when it is above the threshold. The threshold can be adjusted arbitrarily, and further, it may be described in multiple stages including "slightly effective" instead of two values.

[0098] The proposed addition of explanatory variables 455 extracts and stores indicators that may improve the accuracy of risk detection in addition to the existing explanatory variables from the comments 146 from the respondents of the user response results 14.

[0099] The expression method improvement proposals 456 are extracted and stored as improvement proposals for existing risk expressions from the comments 146 from the respondents of the user response results 14. The risk expression improvement proposals may include, for example, adding new information or explanations, changing the presentation medium, presenting countermeasure proposals, etc.

[0100] <Processing details> The process shown in FIG. 2 will be described in detail below.

[0101] 8 is a flowchart showing an example of processing performed in the data processing / storage unit 31. This processing is the processing performed in step S201 in FIG.

[0102] The data processing and storage unit 31 first converts the data obtained from the subjective physical and mental information collection device 9 into indices according to a predetermined rule as appropriate, and stores the indices in the storage device 4 as subjective physical and mental information 41 (S1101). The subjective physical and mental information 41 includes a user ID 411, a date and time 412, and a response to a medical interview in one record. The medical interview is information related to subjective physical and mental information, and the subjective physical and mental information 41 can include, for example, a response value 413 to a medical interview asking about fatigue and / or a response value 414 to a medical interview asking about the soundness of the most recent sleep. In addition, information on emotions related to the worker's stress, such as irritability, may also be used.

[0103] Next, the data processing and storage unit 31 extracts, collates, and formats information obtained from the work-related information server 8 (S1102). First, the data processing and storage unit 31 extracts incident information 463 and work performance information 464 from the work-related information server 8, and stores them in the external information 46. Next, the data processing and storage unit 31 extracts environmental information 82, behavioral information 83, and attendance information 84 from the work-related information server 8, collates and formats the information into one table, and stores the information in the external information 46 as labor information 462. The data processing and storage unit 31 may also estimate nighttime sleeping hours from the behavioral information 83, and include the estimated information in the labor information 462.

[0104] Next, the data processing and storage unit 31 stores the biological information obtained from the biological measurement information collecting device 10 in the external information 46 (S1103). First, the data processing and storage unit 31 stores the measurement time 4611 of the biological information, the average heartbeat interval 4612 measured by the heart rate meter 101, the blood pressure 4615 measured by the sphygmomanometer 103, the body temperature 4616 measured by the thermometer 102, and the arterial blood oxygen saturation 4617 measured by the pulse oximeter 104. Furthermore, the data processing and storage unit 31 calculates TP 4613 and LF / HF 4614 as autonomic nervous function indexes from the heartbeat interval (RRI data) measured by the heart rate meter 101, and stores them in the biological information 461.

[0105] The calculation of TP4613 and LF / HF4614 is performed as follows.

[0106] Fig. 9 is a graph showing an example of heart rate data. The data processing and storage unit 31 calculates heart rate interval data (RRI data) for an analysis window (predetermined period) ΔTw as heart rate variability time series data from the preprocessed heart rate interval data 13 shown in Fig. 9, and further calculates fluctuations from the heart rate variability time series data.

[0107] 10 is a graph showing an example of fluctuations (heart rate variability) in heart rate interval data calculated by the data processing and storage unit 31. The RRI of heart rate interval data is not constant, but fluctuates due to autonomic nerve activity and the like.

[0108] The data processing and storage unit 31 performs frequency spectrum analysis on the heart rate variability time series data to calculate power spectral density (PSD). A known method may be used to calculate the power spectral density.

[0109] Next, the data processing / storage unit 31 calculates the intensity LF of the low frequency component and the intensity HF of the high frequency component of the power spectrum density.

[0110] 11 is a graph showing an example of the frequency domain of the power spectral density of heart rate variability. As shown in FIG. 11, the data processing / storage unit 31 calculates the total autonomic nervous power, i.e., TP4613, as the sum (LF+HF) of the intensity (integral value) LF of the low-frequency component region (0.05 Hz to 0.15 Hz) of the power spectrum and the intensity (integral value) HF of the high-frequency component region (0.15 Hz to 0.40 Hz).

[0111] The data processing and storage unit 31 also calculates the ratio of the intensity LF of the low frequency component of the power spectrum to the intensity HF of the high frequency component (autonomic nerve LF / HF) as LF / HF4614.

[0112] Through the above process, the in-work labor support server 1 calculates heart rate variability time series data for each analysis window ΔTw from the heart rate interval data of the biological information 461, and calculates it as TP 4613 and LF / HF 4614.

[0113] High-frequency components of the power spectral density of heart rate variability appear in heart rate variability when the parasympathetic nervous system is activated (tension), and low-frequency components appear in heart rate variability when both the sympathetic nervous system is activated (tension) and the parasympathetic nervous system is activated (tension).

[0114] It is known that when the sympathetic nervous system is activated, the worker is in a stressed state, and when the parasympathetic nervous system is activated, the worker is in a relaxed state. The sum and ratio of the low-frequency component intensity LF and the high-frequency component intensity HF can be used to determine whether the worker is in a stressed or relaxed state. TP, the sum of the low-frequency component intensity LF and the high-frequency component intensity HF, represents the overall activity of the autonomic nervous system, and a decrease in TP represents the accumulation of long-term fatigue. LF / HF, the ratio of the low-frequency component intensity LF to the high-frequency component intensity HF, represents the balance of the autonomic nervous system, and an increase in LF / HF represents an increase in short-term fatigue.

[0115] In this way, the data processing and storage unit 31 converts and formats various types of measurement data. In particular, by including objective fatigue-related information (objective physical and mental information) such as TP4613 and LF / HF4614 in the biological information, it is expected that the accuracy of labor risk analysis will be improved.

[0116] 12 is a flowchart showing an example of processing performed in the time-series data extraction unit 32. This processing is the processing performed in step S202 in FIG.

[0117] The time-series data extraction unit 32 first refers to the index name and time duration described in the explanatory variables 432, 433 of each rule in the detection rule setting information 43 (S1201). Next, the time-series data extraction unit 32 extracts data of the identified index and the identified time duration for each rule from the subjective mind-body information 41 and external information 46 acquired in step S201 of Fig. 2, and stores the extracted time-series data 1201 in the storage device 4 (S1202).

[0118] 13 is a flowchart showing an example of processing performed by the risk detection unit 33. This processing is processing performed in step S203 of FIG.

[0119] The risk detection unit 33 applies the set detection rules to the extracted time-series subjective physical and mental information and other external information to detect labor risks (S1301). The detection results are stored as labor risk information 42.

[0120] In addition, the labor risk information 42 can be freely set to risks related to the safety of workers or managers in the work and future labor. For example, the probability of such events occurring may be defined as labor risk using the number of near misses 4634, whether an accident has occurred 4635, and whether an industrial accident has occurred 4636 in the incident information 463. Alternatively, if it is considered that mistakes during work lead to impatience or stress and affect safety in the work or future labor problems, the probability that the presence or absence of delays 4642 in the work performance information 464 will be Yes (i.e., a delay will occur) may be defined as labor risk. Alternatively, the probability that TP 4613 and LF / HF 4614 in the subjective physical and mental information 41 or biometric information 461 will decrease in the future may be defined as labor risk.

[0121] In a method of determining labor risk information 42, when incident information 463 or work performance information 464 is considered a risk, the occurrence of an event recorded in incident information 463 or work performance information 464 is used as the correct answer data for the objective variable, multiple indicators related to it, such as subjective physical and mental information 41 and labor information 462, are selected as explanatory variables, and estimation can be performed using statistical analysis, machine learning, or the like using past accumulated data. When a decline in TP4613 or LF / HF4614 of subjective physical and mental information 41 or biological information 461 is considered a risk, past changes in subjective physical and mental information 41 or biological information 461 over time can be added to the explanatory variables and estimation can be performed using statistical analysis, machine learning, or the like. Furthermore, particularly when a decline in subjective physical and mental information 41 is considered a risk, any numerical value can be used as the threshold for correct answer data.

[0122] 16 is a flowchart showing an example of processing performed by the presentation content setting unit 34. This processing is the processing performed in step S204 in FIG.

[0123] When a labor risk is detected by the risk detection unit 33, the presentation content setting unit 34 determines the presentation content 1601 using the labor risk information 42 and the risk expression information 44 (S1601).

[0124] The presentation content setting unit 34 refers to the rule ID 431 of the risk detected in the labor risk information 42, and extracts the corresponding risk expression from the risk expression information 44. The presentation content setting unit 34 determines the presentation content based on the expression method described in the risk expression information 44. The presentation method can be selected from text, graphs, music, light, etc. The presentation content setting unit 34 can also refer to the explanatory variables set in the detection rule setting information 43 and include them in the presentation text. The presentation method text may be displayed as text on the screen or spoken aloud.

[0125] The presentation message 444 can be mechanically determined by preparing multiple sentence templates, such as "{Explanatory variable} is declining, so please be careful of the risk of {Objective variable}." Alternatively, machine learning such as generative AI may be used to output the presentation message 444 in a more flexible and easy-to-understand manner.

[0126] The processing of the alert presentation unit 35 is performed in step S205 of Fig. 2. The presentation content as shown in Fig. 18 determined by the presentation content setting unit 34 is presented on the prediction result display / input terminal carried by the worker or manager.

[0127] Figure 18 is a graph showing an example of labor risk detection result information presented by the system. The labor risk detection result information may be similar to graph 2502 shown in Figure 25, for example. As will be described later, graph 2502 displays the number of near misses and medical interview responses (for example, responses regarding the degree of fatigue) for a certain worker on each working day.

[0128] 19 is a flowchart showing an example of processing performed by the detection and presentation effect acquisition unit 36. This processing is the processing performed in step S206 in FIG.

[0129] The detection and presentation effect acquisition unit 36 ​​asks the worker or manager who has been presented with a labor risk how effective the past detection results and presentation contents were for the user, using a method such as a questionnaire, and acquires the user's response result 14 (S1901). The worker responds to the alert presented to them. The manager can respond to alerts presented to multiple workers, regardless of whether they are under his or her management. However, when the manager responds, he or she must conceal any personally identifiable information, such as the worker's name or facial photograph. This makes it possible to collect information without affecting the business relationship between the worker and the manager. The questions to be asked are distinguished by the detection rule ID 431 and the type of person to whom the alert is presented (in this embodiment, there are two types: worker and manager). This makes it possible to evaluate the effectiveness of the alert on the worker and the alert on the manager, respectively.

[0130] Next, the detection and presentation effect acquisition unit 36 ​​tally the response results 14 from users acquired in step S1901, and acquires detection and presentation effect information 45 indicating whether the risks presented to the workers and managers and the way in which they were presented were appropriate (S1902). The detection and presentation effect acquisition unit 36 ​​distinguishes between the information acquisition date, the detection rule ID 431, and the type of person to whom the information was presented. The detection and presentation effect acquisition unit 36 ​​tally the response results acquired from each unit, and determines the percentage of responses that indicated that the information was effective, the need for improvement, proposed countermeasures, etc.

[0131] Possible countermeasures include a proposal to add explanatory variables 455 and a proposal to improve the expression method 456. The countermeasures may be decided through discussion between administrators and system developers. Alternatively, keywords may be extracted for each detection rule from the comments in the user response results 14, and those that appear most frequently may be selected to mechanically set a proposal to add explanatory variables 455 and a proposal to improve the expression method 456. Alternatively, machine learning such as generation AI may be used to summarize the overall consensus from a large number of comments and use it as a countermeasure.

[0132] 22 is a flowchart showing an example of processing performed by the detection rule setting update unit 37. This processing is part of the processing performed in step S207 in FIG.

[0133] The detection rule setting update unit 37 first refers to the detection rule setting information 43 and the detection and presentation effect information 45 obtained by the detection and presentation effect acquisition unit 36, and if the improvement necessity 454 is "yes" or a value equivalent thereto, extracts candidates for explanatory variables listed in the additional explanatory variable proposal 455 from the subjective physical and mental information 41 and external information 46 (S2201). This makes it possible to select possible indicators that will improve the effectiveness of labor risk detection.

[0134] Next, for each rule in the detection rule setting information 43, the detection rule setting update unit 37 analyzes the relationship between the information extracted in S2201 in addition to the existing explanatory variables and the objective variable 434, and determines whether there is a relationship such as a correlation (S2202). Statistical analysis or machine learning can be used for the analysis.

[0135] If it is determined in step S2202 that there is a relationship between the new group of explanatory variables and the target variable, the detection rule setting update unit 37 updates the existing detection rule setting information 43, adds the extracted information to the explanatory variables, and creates an updated detection rule setting 2201 (S2203).

[0136] The detection rule setting update unit 37 may not only add new explanatory variables but also remove unnecessary explanatory variables. In this case, when extracting keywords from the comments in the user answer results 14, the detection rule setting update unit 37 extracts negative words such as "unnecessary," "disturbing," and "hard to see" and their surrounding words, and describes a proposal for removing explanatory variables in the same way as the proposal for adding explanatory variables 455 in the detection and presentation effect information 45. Furthermore, by deleting the explanatory variables described in the proposal for removal in step S2203 and performing analysis, it is possible to verify the appropriateness of removing the explanatory variables.

[0137] 23 is a flowchart showing an example of processing performed by the risk expression update unit 38. This processing is part of the processing performed in step S208 in FIG.

[0138] The risk expression update unit 38 first extracts explanatory variables added in the updated detection rule settings 2201 updated by the detection rule setting update unit 37 and improvement proposals 456 for the expression method in the detection and presentation effect information 45 (S2301).

[0139] Next, the risk expression update unit 38 determines whether or not to reflect the extracted improvement proposals 456 for the expression methods in the presented content by conducting a re-questionnaire to the users or by the judgment of the administrator or system administrator (S2302). Alternatively, without judging whether or not to do so, it may adopt a policy of adopting all of the proposed improvement proposals 456 for the expression methods, and after presenting them to the user again, the detection and presentation effect acquisition unit 36 ​​may evaluate the effect of the proposals and determine whether to retain or reject the proposals.

[0140] For example, the risk expression update unit 38 may cancel the update if the evaluation of effectiveness input by the user when presenting to the user risk expression information 44 updated in accordance with the proposed improvement 456 for the expression method is lower than the evaluation of the effectiveness of the presentation based on the risk expression information 44 before the update, or may maintain the update if the evaluation is not lower.

[0141] Next, the risk expression update unit 38 generates updated risk expression information 2301 by reflecting the adopted improvement proposal 456 of the expression method in the existing risk expression information 44 (S2303).

[0142] FIG. 24 is a diagram showing an example in which the risk expression information is updated for each user by the risk expression update unit 38, that is, updated according to the presentation target person such as a worker or a manager.

[0143] The risk expression information 500 before update shown in FIG. 24 is an example of the risk expression information 44 before being updated by the risk expression update unit 38. That is, the rule ID 501, the presentation target 502, the presentation information 503, the presentation method 504, and the presentation wording 505 correspond to the rule ID 431, the presentation target 441, the presentation information 442, the presentation method 443, and the presentation wording 444, respectively. The risk expression information 500 illustrated in FIG. 24 is the same as the risk expression information 44 illustrated in FIG. 17. On the other hand, the risk expression information 510 is an example of the risk expression information 44 before being updated by the risk expression update unit 38. That is, the rule ID 511, the presentation target 512, the presentation information 513, the presentation method 514, and the presentation wording 515 correspond to the rule ID 431, the presentation target 441, the presentation information 442, the presentation method 443, and the presentation wording 444, respectively.

[0144] For example, before the update, the risk expression related to the detection rule with rule ID=R1 in the detection rule setting information 43 had the same content of the risk expression (i.e., presented information 503, presentation method 504, and presented message 505) whether the presentation target 502 was a worker or a manager. That is, whether the presentation target 502 was a worker or a manager, the presented information 503 was "objective variable Y, explanatory variables X1, X2," the presentation method 504 was "message," and the presented message 505 was "In the last two weeks, the change in {X2} has been small, while the change in {X1} has been large. Please be careful of the risk of {Y}."

[0145] In this example, as shown in Fig. 14, the explanatory variable X1 of rule ID=R1 is the standard deviation of TP over two weeks, the explanatory variable X2 is the standard deviation of LF / HF over two weeks, and the objective variable Y is the risk of operational mistakes. For this reason, the wording actually presented in accordance with the presented wording 505 is, for example, "In the last two weeks, there has been little change in LF / HF, while there has been a large change in TP. Please be careful of the risk of operational mistakes."

[0146] In this example, checking the user's response results 14 and the risk presentation effect for R1 in the detection and presentation effect information 45 reveals that the workers who were presented with the risk expression before the update were hardly convinced, and that the reason for this was insufficient explanation of the word "TP," which was presented as an explanatory variable, and suggested improving the expression by adding the meaning of the explanatory variable and specific countermeasures. As a result, in the updated risk expression information 510, when the presentation target person 512 is a worker, the presentation message 515 further includes a message explaining that "TP," which is the explanatory variable X1, is an indicator of chronic fatigue, and a message suggesting taking frequent breaks as a countermeasure against chronic fatigue.

[0147] On the other hand, it was also determined that there was a need for improvement when the same risk expression was presented to the manager, but it became clear that there was a stronger desire to be able to refer to labor information at the same time as the detection results, and a proposal to improve the expression was made to display it in a graph to easily show multiple explanatory indicators.As a result, in the updated risk expression information 510, when the presentation target 512 is a worker, the presentation information 513 further includes "number of consecutive attendances," the presentation method 514 further includes "graph," and the presentation text 515 further includes text indicating the number of consecutive attendances.

[0148] As such, the information required from an alert and the way in which it is presented may differ depending on the user's position. In the example in Figure 24, the above-mentioned improvement proposals for expression are adopted for both workers and managers, and it is expected that the sense of satisfaction of the alert will improve for each user.

[0149] Furthermore, with regard to the detection rule with rule ID=R2 in the detection rule setting information 43, when the presentation target person 502 is a "worker," the presentation information 503 is "objective variable Y, explanatory variable X1," the presentation method 504 is "text, graph," and the presentation text 505 is "{X1} has dropped significantly since your last attendance. Are you taking time off to rest? Please be aware of the risk of {Y}."

[0150] In this example, as shown in Figure 14, the explanatory variable X1 of rule ID=R2 is the difference between medical interview 1 (i.e., the answer to the question about how well you are) over two work days, and the objective variable Y is the risk of a near miss. Therefore, the wording that is actually presented in accordance with the presented wording 505 might be, for example, "The answer value to the medical interview about fatigue has dropped significantly since your last work day. Are you taking enough rest time? Please be careful of the risk of a near miss" (see Figure 25).

[0151] In this example, the response results 14 from users to this presentation message and the detection and presentation effect information 45 revealed that the detection risk of R2 was highly convincing to the workers. One factor contributing to this was a comment 146 from the respondent that a large short-term change over two days in the interview response value set as the explanatory variable X1 is likely to occur when the worker's arrival time suddenly changes, such as working an early shift the day after a late shift. Furthermore, the detection rule setting update unit 37 determined that adding an explanatory variable related to arrival time was effective. As a result, in the updated risk expression information 510, the presentation information 513 further includes "delayed arrival time," and the presentation message 515 further includes a statement indicating a large deviation in arrival time and a statement suggesting that the worker get sufficient sleep as a countermeasure.

[0152] FIG. 25 is a diagram showing an example of an alert for a worker presented by the alert presentation unit 35 and an example of an information collection medium for acquiring the alert presentation effect by the detection and presentation effect acquisition unit 36. Here, the focus is on the detection rule (labor risk = near-miss risk) with rule ID = R2 in the detection rule setting information 43. The alert screen 2500 can present a message 2501 and a graph 2502 informing that a labor risk has been detected. The message 2501 contains the presentation wording 505 described in the risk expression information 500 before the update. The graph 2402 can plot fluctuations in the indicators related to the explanatory variables and target variables presented in the message 2501.

[0153] Furthermore, the questionnaire screen 2510 presents a question 2511 for obtaining answer results 14 from the user in order to calculate the detection and presentation effect information 45. This corresponds to the question 144 in the answer results 14 from the user. The format of the answer column 2512 to this question 2511 may be, for example, a VAS format, or a format in which the answer is given in two values, yes / no, or in a multi-level scale.

[0154] Furthermore, an additional question 2513 asking about the reason for the dissatisfaction can be presented to a worker who has answered that they are convinced of the alert. An answer 2514 to the additional question is stored in the comment 146 from the respondent in the user's answer result 14. On the other hand, an additional question (not shown) asking about the reason for the dissatisfaction can be presented to a worker who has answered that they are not convinced of the alert in a format similar to the additional question 2513 and answer 2514 described above, and an answer (not shown) to that question can be accepted. The answer to this additional question is also stored in the comment 146 from the respondent in the user's answer result 14.

[0155] Whether or not the alert is convincing is determined based on the value in the answer field 2512 to this question 2511 if it is a binary yes / no, and if the answer is in VAS format or on a multi-level scale, it can be determined based on whether the answer leans towards convincing / not convincing.

[0156] Although examples of alerts and questionnaires for workers are shown here, alerts and questionnaires for managers can also be presented in the same way.

[0157] 26 is a diagram showing an example of an updated alert for a worker that is presented by the alert presenter 35 after the risk expression has been updated by the risk expression updater 38. Compared to the alert before the risk expression update in FIG. 25, a message 2601 has been added with explanatory information 2602 about a new indicator, deviation in start time, and information 2603 about measures to address the risk. Also, a graph 2604 has been added with additional information 2605 about a change in start time.

[0158] The reason for this update is that, as explained with reference to FIG. 24, the detection risk of R2 is highly acceptable to workers, and as a factor behind this, comment 146 from the respondent pointed out that when there is a large short-term change in the interview response value over two days, this is likely to occur due to a sudden change in the start time of work, and furthermore, the detection rule setting update unit 37 determined that adding an explanatory variable related to the start time of work was effective.

[0159] As described above, the in-work labor support server 1 of this embodiment extracts, organizes, and converts the worker's subjective physical and mental information, work-related information, and biometric information, and stores them as subjective physical and mental information 41 and external information 46.

[0160] This makes it possible to use the worker's subjective physical and mental information, as well as other work-related information, to analyze data stratified by the worker's working conditions and to define and detect labor risks from various perspectives.

[0161] In addition, by setting detailed detection rule setting information and risk expression information for detected labor risks, it is possible to present alerts to workers and managers that are appropriate for their individual situations.

[0162] In addition, by obtaining and analyzing the detection and presentation effect information 45 for the presented alert to the worker or manager and updating the detection rule setting information 43 as necessary, more reliable labor risk detection becomes possible.

[0163] In addition, by obtaining and analyzing the detection and presentation effect information 45 for the presented alert and sending it to the worker or manager, and updating the risk expression information 44 as necessary, it becomes possible to present more convincing alerts that take into account the individual's position.

[0164] <Conclusion> The above embodiment can be configured as follows, for example.

[0165] (1) An information processing system, comprising a processor (e.g., processor 2) and a storage device (e.g., memory 3 and storage device 4), wherein the storage device stores past subjective physical and mental information (e.g., subjective physical and mental information 41) input by a first user (e.g., worker) regarding his or her own physical and mental state at any given time, detection rules (e.g., detection rule setting information 43) for detecting labor risks related to the first user's physical and mental state based on the subjective physical and mental information, and risk expression information (e.g., risk expression information 44) that sets content to be presented to the first user and / or content to be presented to a second user (e.g., manager) different from the first user based on the detected labor risks. 44), and the processor executes a first procedure (e.g., step S202) of extracting the subjective physical and mental information for a predetermined period based on the detection rule; a second procedure (e.g., step S203) of detecting the labor risk based on statistics of the subjective physical and mental information for the predetermined period and the detection rule; a third procedure (e.g., step S204) of generating content to be presented to the first user and / or content to be presented to the second user regarding the detected labor risk based on the risk expression information; and a fourth procedure (e.g., step S205) of outputting the generated content to be presented to the first user and the second user to whom the content is to be presented.

[0166] This makes it possible to present labor risk detection results that are reliable and convincing to users in different positions, such as workers who are the targets of labor risk detection based on subjective physical and mental information, and the managers who manage those workers.

[0167] (2) In the information processing system described in (1) above, the processor executes a fifth step (e.g., step S206) of acquiring information indicating an evaluation of the effectiveness of the presented content from the first user and / or the second user, a sixth step (e.g., step S207) of updating the detection rule based on the evaluation of the effectiveness of the presented content, and a seventh step (e.g., step S208) of updating the risk expression information based on the evaluation of the effectiveness of the presented content.

[0168] This increases the reliability of labor risk detection and the satisfaction of each user with the presentation of labor risks.

[0169] (3) An information processing system as described in (2) above, wherein the storage device stores external information (e.g., external information 46) that is information about the first user other than the subjective physical and mental information, the detection rules include rules for detecting the labor risk based on the subjective physical and mental information and the external information, and the processor further extracts the external information for the specified period in the first step, and detects the labor risk based on statistics of the subjective physical and mental information for the specified period, the external information, and the detection rules in the second step.

[0170] This allows for highly reliable labor risk detection based on not only subjective physical and mental information but also objective physical and mental information.

[0171] (4) An information processing system as described in (3) above, wherein the external information includes at least one of labor information of the first user (e.g., labor information 462), incident information related to business (e.g., incident information 463), and business performance information (e.g., business performance information 464).

[0172] This allows for highly reliable labor risk detection based on not only subjective physical and mental information but also objective physical and mental information.

[0173] (5) An information processing system according to (4) above, wherein the external information includes either biometric information of the first user (e.g., biometric information 461) or objective physical and mental information calculated from the biometric information (e.g., TP4613, LF / HF4616).

[0174] This allows for highly reliable labor risk detection based not only on subjective physical and mental information but also on objective physical and mental information that indicates the user's physical and mental state.

[0175] (6) An information processing system according to (5) above, wherein the biometric information includes heart rate data of the first user, and the objective physical and mental information includes an autonomic nervous function index (e.g., TP4613, LF / HF4616) based on the heart rate data.

[0176] This allows for highly reliable labor risk detection based not only on subjective physical and mental information but also on objective physical and mental information that indicates the user's physical and mental state.

[0177] (7) An information processing system according to (6) above, wherein the detection rule is a rule for calculating the magnitude of the risk based on the subjective physical and mental information or a combination of the subjective physical and mental information and information on at least one item included in the external information.

[0178] This allows for highly reliable labor risk detection based on subjective physical and mental information or a combination of subjective and objective physical and mental information.

[0179] (8) In the information processing system described in (7) above, the detection rule is a rule (e.g., a judgment formula 435 and a threshold 436) that uses at least the values ​​of items included in the subjective physical and mental information as explanatory variables and a labor risk value related to the physical and mental state as a dependent variable, and in the sixth step, the processor analyzes, by statistical analysis or machine learning, the relationship between the explanatory variables and the dependent variable when either an item included in the subjective physical and mental information or an item included in the external information is added as an explanatory variable of the detection rule, or when either an explanatory variable included in the detection rule is deleted, and if the relationship satisfies a predetermined condition, updates the detection rule to reflect the addition or deletion of the explanatory variable.

[0180] This can increase the reliability of labor risk detection.

[0181] (9) In the information processing system described in (2) above, in the fifth step, the processor aggregates, for each acquiring user, evaluations of the effectiveness of the presented content obtained from the plurality of first users and / or evaluations of the effectiveness of the presented content obtained from the plurality of second users, and determines whether or not to update the content of the risk expression information for each acquiring user based on the proportion of positive evaluations.

[0182] This can increase the sense of satisfaction when the results of labor risk detection are presented.

[0183] (10) In the information processing system described in (2) above, in the seventh step, the processor updates the risk expression information, compares an evaluation of the effectiveness of the presented content based on the updated risk expression information with an evaluation of the effectiveness of the presented content based on the risk expression information before the update, and if the evaluation after the update has decreased, cancels the update of the risk expression information, and if the evaluation after the update has not decreased, maintains the update of the risk expression information.

[0184] This can increase the sense of satisfaction when the results of labor risk detection are presented.

[0185] (11) In the information processing system described in (2) above, in the fifth step, the processor inputs evaluations of the effectiveness of the presented content obtained from the plurality of first users and / or evaluations of the effectiveness of the presented content obtained from the plurality of second users into a machine learning model for each obtained user, thereby obtaining summaries of the evaluations from the plurality of first users and / or summaries of the evaluations from the plurality of second users, and generates explanatory variables to add to the detection rule, explanatory variables to delete from the detection rule, and candidates for updating the risk expression information based on the obtained summaries.

[0186] This can increase the reliability of labor risk detection and the persuasiveness of the results of labor risk detection.

[0187] (12) In the information processing system described in (1) above, the subjective physical and mental information is information based on at least one of the first user's response to a question about fatigue (e.g., response value 413) and the first user's response about sleep (e.g., response value 414).

[0188] This allows appropriate subjective physical and mental information to be collected.

[0189] (13) In the information processing system described in (1) above, the first user is an employee of a predetermined business, and the second user is an administrator of the first user.

[0190] This allows for proper management of labor risks in business operations.

[0191] The present invention is not limited to the above-described embodiments and includes various modifications. For example, if a clear correlation between work-related risks and health-related risks is established, the present invention can be applied as a technology to support health management of employees, etc. Furthermore, the technology of the present invention can also be applied to users, such as the relationship between an athlete and his / her manager. In general, the above-described embodiments have been described in detail to clearly explain the present invention and are not necessarily limited to including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of another embodiment. Furthermore, the addition, deletion, or substitution of part of the configuration of each embodiment with other configurations can be applied alone or in combination.

[0192] Furthermore, the above-described configurations, functions, processing units, and processing means may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations and functions may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function may be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0193] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and do not necessarily show all the control lines and information lines in the product. In reality, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0194] 1. In-house labor support server 2 processors 3. Memory 4 Storage device 5. Communication Interface 6 Output Devices 7 Input Devices 8 Business-related information server 9. Subjective mind-body information collection device 10. Biometric information collection device 11 Prediction result display and input terminal 12 Network

Claims

1. An information processing system, a processor and a storage device, The storage device holds past subjective physical and mental information input by a first user regarding his or her physical and mental state at any given time, detection rules for detecting labor risks related to the first user's physical and mental state based on the subjective physical and mental information, and risk expression information that sets content to be presented to the first user and / or content to be presented to a second user different from the first user based on the detected labor risks, The processor: a first step of extracting the subjective psychosomatic information for a predetermined period based on the detection rule; a second step of detecting the labor risk based on the statistics of the subjective mental and physical information for the predetermined period and the detection rule; a third step of generating presentation content for the first user and / or presentation content for the second user regarding the detected labor risk based on the risk expression information; and a fourth step of outputting the generated presentation content to the first user and / or the second user to whom the content is to be presented.

2. 2. The information processing system according to claim 1, The processor: a fifth step of acquiring information indicating an evaluation of an effect of the presented content from the first user and / or the second user; a sixth step of updating the detection rule based on an evaluation of the effectiveness of the presentation content; and a seventh step of updating the risk expression information based on an evaluation of the effectiveness of the presented content.

3. 3. The information processing system according to claim 2, the storage device stores external information that is information about the first user other than the subjective mind-body information; The detection rules include rules for detecting the labor risk based on the subjective physical and mental information and the external information, The processor: In the first step, the external information for the predetermined period is further extracted; An information processing system characterized in that in the second step, the labor risk is detected based on statistics of the subjective physical and mental information for the specified period, the external information, and the detection rule.

4. 4. The information processing system according to claim 3, An information processing system, wherein the external information includes at least one of labor information of the first user, incident information related to work, and work performance information.

5. 5. The information processing system according to claim 4, An information processing system, wherein the external information includes either biometric information of the first user or objective mind-body information calculated from the biometric information.

6. 6. The information processing system according to claim 5, the biological information includes heart rate data of the first user; An information processing system, wherein the objective mind-body information includes an autonomic nervous function index based on the heart rate data.

7. 7. The information processing system according to claim 6, An information processing system characterized in that the detection rule is a rule that calculates the magnitude of the labor risk based on the subjective mental and physical information or a combination of the subjective mental and physical information and information on at least one item included in the external information.

8. 8. The information processing system according to claim 7, The detection rule is a rule that uses at least the values ​​of items included in the subjective mental and physical information as explanatory variables and a labor risk value related to the mental and physical state as a response variable, In the sixth step, the processor analyzes, by statistical analysis or machine learning, the relationship between the explanatory variables and the objective variable when either an item included in the subjective physical and mental information or an item included in the external information is added as an explanatory variable of the detection rule, or when either an explanatory variable included in the detection rule is deleted, and if the relationship satisfies a predetermined condition, updates the detection rule to reflect the addition or deletion of the explanatory variable.

9. 3. The information processing system according to claim 2, In the fifth step, the processor aggregates the evaluations of the effectiveness of the presented content obtained from multiple first users and / or the evaluations of the effectiveness of the presented content obtained from multiple second users for each acquiring user, and determines whether to update the content of the risk expression information for each acquiring user based on the proportion of positive evaluations.

10. 3. The information processing system according to claim 2, In the seventh step, the processor updates the risk expression information, compares an evaluation of the effectiveness of the presented content based on the updated risk expression information with an evaluation of the effectiveness of the presented content based on the risk expression information before the update, and if the evaluation after the update decreases, cancels the update of the risk expression information, and if the evaluation after the update does not decrease, maintains the update of the risk expression information, characterized by the information processing system.

11. 3. The information processing system according to claim 2, In the fifth step, the processor inputs evaluations of the effectiveness of the presented content obtained from a plurality of the first users and / or evaluations of the effectiveness of the presented content obtained from a plurality of the second users into a machine learning model for each obtained user, thereby obtaining summaries of the evaluations from the plurality of the first users and / or summaries of the evaluations from the plurality of the second users, and based on the obtained summaries, generates explanatory variables to add to the detection rule, explanatory variables to delete from the detection rule, and candidates for updating the risk expression information.

12. 2. The information processing system according to claim 1, An information processing system, wherein the subjective physical and mental information is information based on at least one of the first user's response to a medical question about fatigue and the first user's response about sleep.

13. 2. The information processing system according to claim 1, An information processing system, wherein the first user is an employee of a predetermined business, and the second user is an administrator of the first user.

14. An information processing method executed by an information processing system, the information processing system includes a processor and a storage device; The storage device holds past subjective physical and mental information input by a first user regarding his or her physical and mental state at any given time, detection rules for detecting labor risks related to the first user's physical and mental state based on the subjective physical and mental information, and risk expression information that sets content to be presented to the first user and / or content to be presented to a second user different from the first user based on the detected labor risks, The information processing method includes: a first step in which the processor extracts the subjective mind-body information for a predetermined period based on the detection rule; a second step in which the processor detects the labor risk based on statistics of the subjective mental and physical information for the predetermined period and the detection rule; a third step in which the processor generates, based on the risk expression information, presentation content for the first user and / or presentation content for the second user regarding the detected labor risk; and a fourth step in which the processor outputs the generated presentation content to the first user and / or the second user to whom the content is to be presented.

15. An information processing program for controlling an information processing system, the information processing system includes a processor and a storage device; The storage device holds past subjective physical and mental information input by a first user regarding his or her physical and mental state at any given time, detection rules for detecting labor risks related to the first user's physical and mental state based on the subjective physical and mental information, and risk expression information that sets content to be presented to the first user and / or content to be presented to a second user different from the first user based on the detected labor risks, The information processing program a first step of extracting the subjective psychosomatic information for a predetermined period based on the detection rule; a second step of detecting the labor risk based on the statistics of the subjective mental and physical information for the predetermined period and the detection rule; a third step of generating presentation content for the first user and / or presentation content for the second user regarding the detected labor risk based on the risk expression information; and a fourth step of outputting the generated presentation content to the first user and / or the second user to whom the content is to be presented.

Images