WO2024219300A1

WO2024219300A1 - Engagement estimation method, program, and engagement estimation system

Info

Publication number: WO2024219300A1
Application number: PCT/JP2024/014490
Authority: WO
Inventors: 健一入江; 洋介井澤; 若正清崎; 拓磨白井
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2023-04-19
Filing date: 2024-04-10
Publication date: 2024-10-24

Abstract

The purpose of the present disclosure is to reduce the burden on a worker required to estimate engagement. The present disclosure involves a questioning step in which a plurality of images corresponding one-to-one to a plurality of moods are displayed, and a worker is requested to select an image that matches the worker's mood from among the plurality of images. In a second step, response information is stored that relates to which of the plurality of images was selected at each of a plurality of time points. In a fifth step, the worker's engagement in a second period is estimated on the basis of a plurality of feature quantities in the second period and a regression equation. The plurality of feature quantities include feature quantities based on the response information.

Description

Engagement estimation method, program, and engagement estimation system

The present disclosure relates generally to an engagement estimation method, program, and engagement estimation system, and more specifically to an engagement estimation method, program, and engagement estimation system that estimates a worker's engagement with their work.

There is a known indicator of engagement (also called work engagement or employee engagement) related to workers' work. According to non-patent document 1, engagement is defined in terms of the following two points: (1) commitment to the organization, specifically, affective commitment (emotional attachment to the organization) and continuance commitment (desire to remain with the organization); and (2) extra-role behavior (any behavior that enables the organization to function effectively).

Increased engagement can, for example, lead to increased productivity, sales, customer satisfaction, and worker retention within an organization.

Traditionally, engagement has been quantified by conducting surveys of workers and analyzing their responses (e.g., Patent Documents 1 and 2).

However, it may take a lot of time for workers to answer the questionnaire, or they may find it bothersome. In this way, answering the questionnaire may become a burden for workers.

JP 2018-185680 A JP 2019-175108 A

The present disclosure aims to provide an engagement estimation method, program, and engagement estimation system that can reduce the burden on workers required to estimate engagement.

An engagement estimation method according to one aspect of the present disclosure is a method executed by an engagement estimation system to estimate a worker's engagement with work. The engagement estimation method has a first step, a second step, a third step, a fourth step, and a fifth step. In the first step, a questioning step is carried out at multiple time points. In the questioning step, multiple images corresponding one-to-one to multiple moods are displayed, and the worker is asked to select an image from the multiple images that matches the worker's mood. In the second step, answer information regarding which image was selected from the multiple images at each of the multiple time points is stored. In the third step, multiple feature amounts are determined. In the fourth step, a regression equation is obtained that represents the relationship between the multiple feature amounts in a first period and the engagement of the worker in the first period that has been determined in advance. In the fifth step, the engagement of the worker in the second period is estimated based on the multiple feature amounts in a second period different from the first period and the regression equation. The multiple feature amounts include at least one of a feature amount based on biometric information of the worker measured by a biometric information measuring terminal, a feature amount based on location information of the worker at the worker's workplace, a feature amount based on relationship information regarding a relationship between the worker and another worker in the workplace, and a feature amount based on a usage history of a computer system used by the worker for the work. The multiple feature amounts further include a feature amount based on the response information.

A program according to one aspect of the present disclosure is a program for causing one or more processors of a computer system to execute the engagement estimation method.

An engagement estimation system according to one aspect of the present disclosure estimates a worker's engagement with work. The engagement estimation system includes an acquisition unit, a memory unit, a feature determination unit, a regression unit, and an estimation unit. The acquisition unit acquires answer information from a worker terminal. The worker terminal performs a questioning step at multiple time points. In the questioning step, multiple images corresponding one-to-one to multiple moods are displayed, and the worker is prompted to select an image from the multiple images that matches the worker's mood. The answer information is information regarding which image was selected from the multiple images at each of the multiple time points. The memory unit stores the answer information. The feature determination unit determines multiple feature values. The regression unit acquires a regression equation representing the relationship between the multiple feature values in a first period and the worker's engagement in the first period that has been determined in advance. The estimation unit estimates the engagement of the worker in a second period different from the first period based on the multiple feature amounts in the second period and the regression equation. The multiple feature amounts include at least one of a feature amount based on biometric information of the worker measured by a biometric information measuring terminal, a feature amount based on location information of the worker at the worker's workplace, a feature amount based on relationship information regarding a relationship between the worker and another worker in the workplace, and a feature amount based on a usage history of a computer system used by the worker for the work. The multiple feature amounts further include a feature amount based on the response information.

FIG. 1 is a block diagram of an engagement estimation system and related components according to one embodiment. FIG. 2 is a graph illustrating a process in the engagement estimation system. FIG. 3 is an explanatory diagram showing an engagement estimation process performed by the engagement estimation system. FIG. 4 is an explanatory diagram showing an operation window displayed on a PC in the engagement estimation method using the engagement estimation system. FIG. 5 is a graph showing an example of response information acquired by the engagement estimation system. FIG. 6 is a graph showing an example of the amount of change in response information acquired by the engagement estimation system. FIG. 7 is a flowchart showing a process flow of a part of an engagement estimation method using the engagement estimation system.

(Embodiment)
Hereinafter, an engagement estimation method, a program, and an engagement estimation system 1 according to an embodiment will be described with reference to the drawings. However, the embodiment described below is merely one of various embodiments of the present disclosure. The embodiment described below can be modified in various ways depending on the design, etc., as long as the object of the present disclosure can be achieved.

(overview)
FIG. 1 shows a schematic configuration of an engagement estimation system 1 of this embodiment. The engagement estimation system 1 is used to estimate the engagement of workers. In the present disclosure, "worker" refers to anyone who works in general. Unlike "laborer" in the general sense, "worker" in the present disclosure refers not only to those who receive remuneration in exchange for labor, but also to those who work without remuneration. However, in the following, as a representative example, a case where the engagement of a person who receives remuneration in exchange for labor is estimated will be described. Therefore, the engagement estimation system 1 is used, for example, in companies, government offices, or organizations. In the following, as a representative example, a case where the engagement estimation system 1 is used in a company will be described.

Engagement estimation system 1 of this embodiment shown in FIG. 1 estimates a worker's engagement with work. Engagement estimation system 1 includes a (first) acquisition unit 21, a memory unit 11, a feature determination unit 23, a regression unit 24, and an estimation unit 25. The acquisition unit 21 acquires answer information from a worker terminal (PC 8). The worker terminal performs a question step at multiple time points. In the question step, multiple images corresponding one-to-one to multiple moods are displayed (see FIG. 4), and the worker is prompted to select an image that corresponds to the worker's mood from the multiple images. The answer information is information regarding which image was selected from the multiple images at each of the multiple time points. The memory unit 11 stores the answer information. The feature determination unit 23 determines multiple feature values. The regression unit 24 acquires a regression equation (see FIG. 2) that represents the relationship between the multiple feature values in a first period (see FIG. 3) and the worker's engagement in the first period that has been determined in advance. The estimation unit 25 estimates the worker's engagement in a second period (see FIG. 3) different from the first period, based on a plurality of feature amounts in the second period and a regression equation. The plurality of feature amounts include at least one of a feature amount based on the worker's biometric information measured by the biometric information measurement terminal 3, a feature amount based on the worker's location information in the worker's workplace, a feature amount based on relationship information regarding the relationship between the worker and another worker in the workplace, and a feature amount based on the usage history of a computer system (PC 8) used by the worker for work. The plurality of feature amounts further includes a feature amount based on response information.

In this embodiment, the worker selects an image that fits the worker's mood from among multiple images, and the engagement estimation system 1 estimates the worker's engagement. The worker can intuitively select an image that fits their current mood. This reduces the burden on the worker in answering questions. For example, in this embodiment, the burden on the worker is reduced compared to when a question is displayed, the worker reads the question, and then the worker answers.

Furthermore, according to this embodiment, by using at least one of the worker's biometric information, location information, relationship information, and computer system usage history as a feature, it is possible to estimate engagement more objectively than when engagement is estimated based solely on responses to questions. Furthermore, by using at least one of the worker's biometric information, location information, relationship information, and computer system usage history, it is possible to reduce the number of questions other than those asking about the worker's mood, and to estimate engagement without asking any questions other than those asking about the worker's mood. This reduces the burden on people (workers, etc.) who answer questions.

The engagement estimation method of this embodiment is executed by the engagement estimation system 1, and is a method for estimating a worker's engagement with work. The engagement estimation method has a first step, a second step, a third step, a fourth step, and a fifth step. In the first step, a questioning step is performed at multiple time points. In the questioning step, multiple images corresponding one-to-one to multiple moods are displayed, and the worker is asked to select an image that corresponds to the worker's mood from the multiple images. In the second step, answer information regarding which image was selected from the multiple images at each of the multiple time points is stored. In the third step, multiple feature amounts are determined. In the fourth step, a regression equation is obtained that represents the relationship between the multiple feature amounts in the first period and the worker's engagement in the first period that is determined in advance. In the fifth step, the worker's engagement in the second period is estimated based on the multiple feature amounts in a second period different from the first period and the regression equation. The multiple feature amounts include at least one of the following: a feature amount based on the worker's biometric information measured by the biometric information measuring terminal 3; a feature amount based on the worker's location information at the worker's workplace; a feature amount based on relationship information regarding the relationship between the worker and other workers at the workplace; and a feature amount based on the usage history of a computer system (PC8) used by the worker for work. The multiple feature amounts further include a feature amount based on response information.

The engagement estimation method can also be embodied in a program. The program of this embodiment is a program for causing one or more processors of a computer system to execute the engagement estimation method. The program may be recorded on a non-transitory recording medium that can be read by the computer system.

(detail)
(1) Overall Configuration The engagement estimation system 1 and each of the related components will be described in more detail below.

In this embodiment, there are multiple workers. The engagement estimation system 1 estimates the engagement of each of the multiple workers.

The engagement estimation system 1 is used together with, for example, a bio-information measuring terminal 3, a position measuring system 4, a data server 5, an operation terminal 6, an information processing server 7, a PC (personal computer) 8, an attendance management system 9, and an exercise measuring terminal 10.

(2) Biological Information Measurement Terminal The biological information measurement terminal 3 measures biological information of each of the multiple workers. The biological information includes, for example, at least one of heart rate, blood pressure, skin temperature, sweat rate, and voice information. A single biological information measurement terminal 3 may measure multiple types of biological information (for example, heart rate and blood pressure). Alternatively, there may be multiple biological information measurement terminals 3, and each of the multiple biological information measurement terminals 3 may measure a different type of biological information.

The bioinformation measuring terminal 3 is, for example, a wearable terminal worn by a worker. The wearable terminal is equipped with, for example, an optical heart rate sensor, which measures the worker's heart rate and blood pressure. The wearable terminal is also equipped with, for example, a temperature sensor, which measures the worker's skin temperature. The wearable terminal is also equipped with, for example, a sweat sensor, which measures the amount of sweat produced by the worker.

As another example, the bio-information measuring terminal 3 captures an image of a worker for a certain period of time using a camera (such as a near-infrared camera) to generate image data, and measures the worker's heart rate based on the image data.

As another example, the bio-information measuring terminal 3 is, for example, a blood pressure monitor with an arm band, and measures the worker's blood pressure while the arm band is wrapped around the worker's arm.

As another example, the bioinformation measuring terminal 3 is provided with a microphone, for example, and converts the voice of the worker into audio information in the form of an electrical signal using the microphone. The microphone may be provided in a wearable terminal.

(3) Position Measurement System The position measurement system 4 measures position information of each of the multiple workers. The position information includes, for example, coordinate information of each of the multiple workers.

For example, each of the multiple workers carries a mobile terminal such as a smartphone or a wearable terminal. Multiple beacon devices are installed in the workplaces of the multiple workers (e.g., office buildings, stores, or factories).

Below, we will focus on one worker among multiple workers and explain an example of measuring the location information of this worker. The location information of other workers can also be measured in a similar manner.

Each of the multiple beacon devices transmits a beacon signal. A mobile device carried by the worker measures the received signal strength of the beacon signal. Information on the received signal strength is transmitted from the mobile device to the position measurement system 4. The position measurement system 4 calculates the distance between the mobile device and each of the multiple beacon devices based on the received signal strength. Furthermore, the position measurement system 4 measures the position information of the mobile device by three-point positioning based on the distance between the mobile device and each of the multiple beacon devices and the position information of each of the multiple beacon devices. The position measurement system 4 transmits the position information of the mobile device to the engagement estimation system 1 as position information of the worker carrying the mobile device.

The mobile terminal (e.g., a wearable terminal) that receives the beacon signal may also serve as the biological information measurement terminal 3.

(4) Data Server The data server 5 stores relationship information. The relationship information is information about the relationships between the multiple workers in their workplaces (such as a company, a government office, or an organization). More specifically, the relationship information includes, for example, information about the hierarchical relationships between the multiple workers. The information about the hierarchical relationships between the multiple workers is, for example, information about the job titles of each of the multiple workers. The job title refers to a work position. The job title is a position or a rank. The relationship information also includes, for example, organizational information about the organization (department or division, etc.) to which each of the multiple workers belongs. The department or division is distinguished by the name, for example, XX Department, XX Section, or XX Center. The relationship information also includes, for example, business information (for example, information about the name of the business) related to the identification of the business (project, etc.) in which each of the multiple workers is involved.

The data server 5 also stores area information. The area information includes, for example, map information of the workplaces of multiple workers. The area information includes, for example, information on the location of each room and the purpose of each room.

(5) Operation Terminal The operation terminal 6 is, for example, a personal computer or a mobile terminal. The mobile terminal is, for example, a mobile phone such as a smartphone, a wearable terminal, or a tablet terminal.

The operation terminal 6 generates the worker's work-related reporting information in response to human operation. In order to generate the worker's reporting information on the operation terminal 6, the worker himself/herself may operate the operation terminal 6, or another person may operate the operation terminal 6.

The operation terminal 6 includes, for example, a touch panel display, and displays survey items on the touch panel display. A person answers the survey items by operating the touch panel display. The operation terminal 6 then generates declaration information that includes the answers obtained from the person.

For example, when a question is presented as a questionnaire item, a person selects an answer from among multiple options. The multiple options are, for example, five options: "Agree," "Somewhat agree," "Can't say," "Somewhat disagree," and "Disagree."

The questionnaire items include, for example, questions for estimating engagement in the first period described above in the information processing server 7. Such questions are hereinafter referred to as "first questions", and answers to the first questions are hereinafter referred to as "first answers". The first questions are, for example, questions asking about the worker's place of employment, the content of the work, and the worker's thoughts and feelings about their colleagues.

The questionnaire items also include, for example, questions for estimating engagement in a second period different from the first period by the engagement estimation system 1. Such questions are hereinafter referred to as "second questions," and answers to the second questions are hereinafter referred to as "second answers." At least one second question may be the same as the first question. It is preferable that the number of items (number of questions) of the second questions is less than the number of items (number of questions) of the first questions.

(6) Information Processing Server The information processing server 7 estimates the engagement of the worker in a first period. More specifically, the information processing server 7 first acquires declaration information from the operation terminal 6. The declaration information includes at least one first response. Based on the at least one first response, the information processing server 7 estimates the engagement of the worker in the first period.

Engagement is expressed, for example, as a numerical value. A known method such as that disclosed in Patent Document 1 can be adopted as a method for the information processing server 7 to estimate engagement. For example, a score for the first answer is determined depending on which of multiple options is selected as the first answer. The information processing server 7 estimates the sum of the scores for each of the multiple first answers as the worker's engagement in the first period.

(7) PC
The workers use PCs (personal computers) 8 for work. More specifically, each of the workers is assigned one or more PCs 8 by his/her workplace. Each of the workers uses the one or more PCs 8 assigned to him/her.

PC8 is equipped with a memory unit 84 as a storage device that stores the usage history of the PC8. The memory unit 84 is a hard disk drive (HDD) or a solid state drive (SSD) or the like. Software for acquiring the usage history of PC8 and storing it in the memory unit 84 may be installed in PC8.

The PC 8 also includes a processing unit 81, a display unit 82, and an operation unit 83.

The processing unit 81 includes a computer system having one or more processors and a memory. The functions of the processing unit 81 are realized by the processor of the computer system executing a program recorded in the memory of the computer system. The program may be recorded in the memory, or may be provided via a telecommunications line such as the Internet, or may be recorded on a non-transitory recording medium such as a memory card and provided.

The display unit 82 has a display for displaying information.

The operation unit 83 accepts operations by a person. The operation unit 83 has at least one of, for example, a mouse, a keyboard, a button, and a touch panel.

The processing unit 81 controls the display unit 82, the operation unit 83, and the memory unit 84. At a predetermined time, the processing unit 81 causes the display unit 82 to display an operation window 820 (see FIG. 4) that displays a question for prompting the worker to select an image from a plurality of images that fits the worker's mood. The worker can answer the question by operating the operation unit 83 to select one of the images. Answer information regarding which image from the plurality of images was selected is stored in the memory unit 84.

　PC8 is pre-installed with software for executing a question step that displays an operation window 820 on the display unit 82 and prompts the worker to select one of the images. The software may be provided to PC8 from engagement estimation system 1 via wired or wireless communication, or may be provided to PC8 from another device.

The processing unit 81 performs the questioning step at multiple time points. More specifically, the processing unit 81 performs the questioning step at multiple time points in the first period and at multiple time points in the second period.

As an example, the multiple time points are two time points, morning and evening. That is, the processing unit 81 performs the questioning step in the morning and evening of the first period, and in the morning and evening of the second period. In more detail, as an example, the multiple time points are the time when the worker arrives at work and the time immediately before the worker leaves work. Thus, in the engagement estimation method of this embodiment, in the first step, the questioning step is performed at multiple time periods in one day.

As an example, the operation window 820 is automatically displayed as a pop-up during specified time periods in the morning and evening. A worker can cause the processing unit 81 to execute processing by moving the mouse cursor and clicking a specified area of the operation window 820.

In the engagement estimation method of this embodiment, each of the multiple images displayed in the first step is an image showing a person's facial expression. That is, as shown in FIG. 4, multiple images showing people's facial expressions are displayed in the operation window 820. More specifically, each of the multiple images is an illustration. In FIG. 4, displayed from left to right are an image showing a very good mood (very good), an image showing a good mood (good), an image showing a bad mood (bad), and an image showing a very bad mood (very bad). In addition, the mood that each image represents is displayed in text near (below) each image.

When the operation window 820 is displayed, the worker clicks on one of the images. The processing unit 81 then determines that the clicked image has been selected and generates answer information.

The answer information is represented by a number: 1, 2, 3, or 4. That is, when an image that represents a very good mood is selected, the answer information is 1. When an image that represents a good mood is selected, the answer information is 2. When an image that represents a bad mood is selected, the answer information is 3. When an image that represents a very bad mood is selected, the answer information is 4.

As shown in FIG. 4, the operation window 820 displays a minimize button 821, a maximize button 822, and a close button 823. When the minimize button 821 is clicked, the operation window 820 is hidden, and when an icon displayed on the task bar is then clicked, the operation window 820 is displayed again. When the maximize button 822 is clicked, the operation window 820 is displayed over the entire screen of the display unit 82. When the close button 823 is clicked, the software for executing the question step is terminated and the operation window 820 is hidden, and when an operation is then performed to restart the software for executing the question step, the operation window 820 is displayed again.

Furthermore, if no image is selected within a certain time after the operation window 820 is displayed on the display unit 82, the processing unit 81 issues a notification urging the selection of an image. The notification is issued, for example, by displaying a message prompting the selection as a pop-up on the display unit 82. In this way, the engagement estimation method of this embodiment has a notification step of issuing a notification prompting the selection if an image that matches the worker's mood is not selected in the first step. Furthermore, the processing unit 81 issues a notification prompting the selection again when a certain time has passed since issuing the notification prompting the selection. In other words, in the engagement estimation method of this embodiment, if an image that matches the worker's mood is not selected in the first step, a notification prompting the selection is issued at certain time intervals until a selection is made.

PC8 is an example of a computer system that a worker uses for work. A computer system includes one or more computers. The computer system that a worker uses for work is not limited to PC8, and may be, for example, a mobile phone such as a smartphone, a tablet terminal, or a host computer. In addition, the computer system that a worker uses for work may be, for example, a computer system for operating an object to be operated, such as a vehicle or a machine tool, and the usage history may include the operation history of the object to be operated.

The PC 8 is also an example of a worker terminal that performs the above-mentioned questioning steps at multiple points in time. The worker terminal is not limited to the PC 8, and may be, for example, a mobile phone such as a smartphone, or a tablet terminal.

In this embodiment, a questionnaire (question) is presented on the operation terminal 6, and the worker or another person answers the questionnaire. As a result, declaration information is generated on the operation terminal 6. Also, in this embodiment, a question is presented on the PC 8 to prompt the worker to select an image that matches the worker's mood from among multiple images, and the worker himself answers the question. As a result, answer information representing the worker's mood is generated on the PC 8. That is, in this embodiment, questions are presented on both the operation terminal 6 and the PC 8. However, as a variation, the PC 8 may also function as the operation terminal 6. That is, the questions presented on the operation terminal 6 in this embodiment may be presented by the PC 8. Alternatively, the questions presented on the PC 8 in this embodiment may be presented by the operation terminal 6.

(8) Attendance Management System The attendance management system 9 generates attendance information for each of a plurality of workers. For example, each of the plurality of workers carries a readable device such as a mobile terminal (such as a smartphone or wearable terminal) or an IC card, and holds the readable device over the reading device of the attendance management system 9 when arriving at and leaving work. The attendance management system 9 then reads the identification information stored in the readable device from the readable device. As a result, the attendance management system 9 generates attendance information for each of a plurality of workers, and the attendance information includes information on the arrival time and the leaving time.

(9) Exercise Measurement Terminal The exercise measurement terminal 10 obtains an exercise index of the worker. The exercise index indicates at least one of the quality and quantity of exercise. The exercise index includes, for example, at least one of the amount of activity and the amount of movement. The amount of activity is expressed, for example, in METs (Metabolic equivalents). The amount of movement is, for example, the number of steps.

The exercise measuring terminal 10 is, for example, a wearable terminal. The worker carries the wearable terminal. The wearable terminal is equipped with, for example, a pedometer and measures the number of steps taken by the worker. The wearable terminal also measures, for example, the worker's biometric information (heart rate, blood pressure, skin temperature, amount of sweat, etc.) as described above. The wearable terminal calculates the worker's activity level based on the biometric information.

(10) Engagement Estimation System The engagement estimation system 1 includes a processing unit 2, a memory unit 11, and a communication unit 12.

The storage unit 11 is a storage device configured with a hard disk drive (HDD) or a solid state drive (SSD) or the like. The storage unit 11 stores information. The storage unit 11 stores information acquired from an external device, such as biometric information acquired from the biometric information measuring terminal 3, location information acquired from the location measuring system 4, relationship information acquired from the data server 5, and usage history and response information acquired from the PC 8.

The communication unit 12 includes a communication interface device. The communication unit 12 is capable of communicating with external devices (e.g., the bioinformation measurement terminal 3, the position measurement system 4, the data server 5, and the PC 8) via the communication interface device. In this disclosure, "capable of communication" means that signals can be sent and received directly or indirectly via a network or a repeater, etc., using an appropriate communication method such as wired communication or wireless communication.

The processing unit 2 includes a computer system having one or more processors and a memory. The functions of the processing unit 2 are realized by the processor of the computer system executing a program recorded in the memory of the computer system. The program may be recorded in the memory, or may be provided via a telecommunications line such as the Internet, or may be recorded on a non-transitory recording medium such as a memory card and provided.

The processing unit 2 has a first acquisition unit 21, a second acquisition unit 22, a feature determination unit 23, a regression unit 24, an estimation unit 25, a presentation content generation unit 26, and a communication processing unit 27. Note that these merely indicate the functions realized by the processing unit 2, and do not necessarily indicate a concrete configuration.

(10.1) First Acquisition Unit The first acquisition unit 21 acquires worker information (information on each of a plurality of workers) via the communication unit 12. The worker information includes the worker's bioinformation measured by the bioinformation measuring terminal 3, the worker's position information measured by the position measuring system 4, and the relationship information stored in the data server 5.

The worker information further includes reporting information generated by the operation terminal 6. The reporting information is information about the work and is generated in response to operations on the operation terminal 6. The reporting information is, for example, one or both of the first and second responses described above.

The worker information also includes the usage history of the computer system that the worker uses for work. In this embodiment, the computer system is PC8. That is, the worker information includes the usage history of PC8.

The worker information also includes answer information that indicates the worker's mood. The answer information is generated by the PC 8 as the worker operates the PC 8.

In addition, the worker information further includes the worker's attendance information. The worker's attendance information is generated by the attendance management system 9.

The worker information further includes the worker's exercise index. The worker's exercise index is generated by the exercise measurement terminal 10.

(10.2) Second Acquisition Unit The second acquisition unit 22 acquires the engagement of a worker in a first period estimated by the information processing server 7.

(10.3) Feature Determining Unit and Regression Unit The feature determining unit 23 determines a plurality of features from the worker information based on the relationship between the worker information and the engagement in the first period. For example, the feature determining unit 23 extracts a plurality of parameters from the worker information. The plurality of parameters are, for example, the duration of a conversation between the worker and a specific person, the amount of activity of the worker, and the overtime hours of the worker.

A certain parameter may coincide with certain information about a worker. In other words, the information about the worker may be used as a parameter directly. For example, the overtime hours of a worker is an example of information about a worker. The overtime hours of a worker may be used as a parameter.

Alternatively, a certain parameter may be found based on certain information about the worker. That is, the information about the worker may be processed to become the parameter. For example, the position information of the worker measured by the position measurement system 4 and the relationship information stored in the data server 5 are each an example of worker information. Based on these, the conversation time between the worker and a specific person may be found as a parameter. More specifically, the conversation time (face-to-face) between the worker and the supervisor can be found, for example, from the position information of the worker and the supervisor. That is, the time when the distance between the worker and the supervisor is within a specified distance (for example, one meter) can be determined as the conversation time between the worker and the supervisor. Also, whether or not a certain worker is a supervisor can be identified based on the relationship information.

The characteristic amount determining unit 23 may determine multiple characteristic amounts for each worker. In other words, to determine multiple characteristic amounts corresponding to a certain worker, information about the worker and the worker's engagement during the first period may be referenced.

Alternatively, the feature determination unit 23 may determine multiple features common to two or more workers. In other words, to determine multiple features corresponding to two or more workers, information on each of the two or more workers and the engagement of each of the two or more workers in the first period may be referenced.

The regression unit 24 acquires a regression equation that expresses the relationship between the multiple feature amounts in the first period and the worker engagement in the first period that was determined in advance based on the first question described above. In the present disclosure, the process of "acquiring a regression equation" may be a process in which the regression unit 24 acquires a regression equation determined in an external configuration of the engagement estimation system 1 from the external configuration, or a process in which the regression unit 24 acquires a regression equation stored in the memory unit 11 of the engagement estimation system 1. Alternatively, the process of "acquiring a regression equation" may be a process of determining a regression equation. In the present embodiment, the regression unit 24 performs the process of "acquiring a regression equation" by performing a process of determining a regression equation based on the multiple feature amounts in the first period and the worker engagement in the first period that was determined in advance.

In this embodiment, as described below, the process in which the feature determination unit 23 determines multiple feature quantities and the process in which the regression unit 24 determines a regression equation are performed in an integrated manner.

Each of the multiple feature quantities is a factor that determines engagement. The multiple parameters are obtained based on worker information, etc. When focusing on a specific parameter among the multiple parameters, the specific parameter may have a relatively strong correlation with engagement, a relatively weak correlation, or no correlation at all.

The feature determination unit 23 determines, among the multiple parameters, a parameter that has a strong correlation with the engagement in the first period as a feature. The feature determination unit 23 calculates the strength of the correlation, for example, by multiple regression analysis. That is, the feature determination unit 23 (and the regression unit 24) obtains a multiple regression equation using the engagement in the first period as the objective variable and the multiple parameters in the first period as multiple explanatory variables, and further obtains a coefficient of determination of the multiple regression equation. The coefficient of determination is a value between 0 and 1. The larger the coefficient of determination, the stronger the correlation. The feature determination unit 23 determines the multiple feature quantities based on the coefficient of determination. For example, the feature determination unit 23 determines the multiple explanatory variables when the coefficient of determination is greater than a threshold value as the multiple feature quantities. Note that the feature determination unit 23 may obtain a simple regression equation instead of the multiple regression equation, and obtain the strength of correlation (coefficient of determination) from the simple regression equation. Alternatively, the feature determination unit 23 may obtain the coefficient of determination using another machine learning model. In short, the feature determination unit 23 can determine the correlation between multiple features and engagement by regression analysis or the like.

Figure 2 illustrates an example of a simple regression equation (straight line L1) obtained with engagement in the first period as the dependent variable and one parameter in the first period as the explanatory variable. Specifically, the explanatory variable is the number of conversations with a second-level supervisor (the direct supervisor's further direct supervisor). The dependent variable and explanatory variable are each obtained for multiple time periods. In Figure 2, pairs of the dependent variable and explanatory variable for each time period are plotted as points. A simple regression equation is obtained based on these multiple points.

When determining multiple features corresponding to one specific worker, Figure 2 will be a plot of data (pairs of objective variables and explanatory variables) for that one specific worker. When determining multiple features common to two or more workers, Figure 2 will be a plot of data (pairs of objective variables and explanatory variables) for each of the two or more workers.

The feature determination unit 23 also determines, as the feature, at least one of the following four parameters: a parameter based on the worker's biometric information measured by the biometric measurement terminal 3, a parameter based on the worker's location information at the worker's workplace, a parameter based on relationship information regarding the relationship between the worker and other workers at the workplace, and a parameter based on the usage history of the computer system (PC8) used by the worker for work.

When the first period includes a first time point and a second time point, the feature determination unit 23 may determine multiple feature amounts based on the difference between the engagement at the first time point and the engagement at the second time point. In other words, the feature determination unit 23 may determine multiple feature amounts based on the amount of change in engagement. For example, the feature determination unit 23 may determine a multiple regression equation using multiple parameters as multiple explanatory variables and the amount of change in engagement as the objective variable, and may determine, from among the multiple parameters, a parameter that has a strong correlation with the amount of change in engagement as the feature amount.

Furthermore, the feature determination unit 23 may set the explanatory variable as the difference between a predetermined parameter and a reference value. The reference value may be, for example, the average value of the predetermined parameter in a specific period. For example, the predetermined parameter is set to the activity level of the worker in a first period. In this case, the reference value may be, for example, the average value of the activity level of the worker in the same period of the previous year to the first period. Alternatively, the reference value may be, for example, the average value of the activity level of the worker from a predetermined number of days (for example, six months) before the first period to the first period. By taking the difference between the predetermined parameter and the reference value, it is possible to reduce the possibility that differences in the reference value for each worker will affect the engagement estimation result.

(10.3.1) Feature Amount Based on Answer Information The multiple feature amounts include at least a feature amount based on answer information that represents the mood of the worker.

The answer information is information regarding which of the multiple images was selected at each of the multiple time points (morning and evening). As mentioned above, the answer information is expressed as a number: 1, 2, 3, or 4.

Figure 5 shows an example of response information for the morning and evening of each day. In the morning and evening of June 5th, and in the evening of June 6th, an image representing a good mood (good) was selected, so the response information is 2. In the morning of June 6th, an image representing a bad mood (bad) was selected, so the response information is 3.

In the engagement estimation method of this embodiment, the third step (a step of determining multiple feature quantities) includes a step of determining the amount of change in the worker's mood between multiple time points. More specifically, in the third step, the feature quantity determination unit 23 determines the amount of change in mood from morning to evening for each day. The amount of change is the value obtained by subtracting the value of the answer information in the morning of the same day from the value of the answer information in the evening. If the worker's mood does not change from morning to evening, the amount of change is 0. If the worker's mood improves from morning to evening, the amount of change is a positive value. If the worker's mood deteriorates from morning to evening, the amount of change is a negative value.

For example, as shown in Figure 5, the response information for the morning and evening of June 5th is 2. Therefore, the change in mood on June 5th is 2-2=0 (see Figure 6). Also, as shown in Figure 5, the response information for the morning of June 6th is 3, and the response information for the evening of the same day is 2. Therefore, the change in mood on June 6th is 2-3=-1 (see Figure 6).

In the engagement estimation method of this embodiment, the feature based on the response information includes the variability in the amount of change in the worker's mood. The variability is the standard deviation σ calculated from the amount of change in the worker's mood over a certain period (here, n days). In other words, when data on the amount of change in mood of a certain worker over a period of n days is obtained, and the amount of change in mood on the 1st, 2nd, 3rd, ..., nth days are X1, X2, X3, ..., Xn, respectively, and the average value of X1 to Xn is Xave, the standard deviation σ is calculated using the following formula.

In this embodiment, the standard deviation σ (variation) calculated in this manner is a feature based on the response information. As a modified example, the variation as a feature may be variance instead of the standard deviation σ.

The correlation between variability and engagement is expressed by a regression equation as described above. A regression equation is calculated for each individual worker. As an example, when a regression equation showing the correlation between variability and engagement is calculated for a first worker, the greater the variability, the lower the engagement. Note that when a regression equation showing the correlation between variability and engagement is calculated for a second worker different from the first worker, the rate of change in engagement relative to a change in variability may be different compared to the first worker, or the greater the variability, the higher the engagement.

(10.4) Estimation Unit The estimation unit 25 executes the fifth step (estimation step). That is, the estimation unit 25 estimates the engagement of the worker in a second period (the most recent period, as an example here) different from the first period, based on the multiple feature amounts determined by the feature amount determination unit 23 and the regression equation obtained by the regression unit 24. For example, it is assumed that three parameters, namely, the conversation time between the worker and the second-level supervisor, the overtime hours of the worker, and the variance in the amount of change in the worker's mood, are determined as feature amounts by the feature amount determination unit 23. In this case, the estimation unit 25 obtains the most recent engagement reflecting the parameters of the previous month by substituting the above three parameters in the previous month into a multiple regression equation (obtained by the regression unit 24) in which each of the above three parameters is an explanatory variable and the engagement is an objective variable.

If parameters are obtained at multiple time points during a second period that is different from the first period, the engagement may be calculated, for example, by substituting the average value of the parameters at the multiple time points into the multiple regression equation. Alternatively, multiple engagements may be calculated by substituting the parameters at the multiple time points into the multiple regression equation.

(10.5) Presentation content generation unit The presentation content generation unit 26 executes a presentation content generation step. That is, the presentation content generation unit 26 generates content to be presented to the viewer based on the estimation result in the fifth step (estimation step) by the estimation unit 25. The content to be presented to the viewer includes, for example, a numerical value representing the engagement obtained in the fifth step. The viewer may be the worker himself/herself who is the target of the engagement request, or may be another person (for example, the worker's supervisor).

The presentation content generation unit 26 generates, for example, a list of the engagement of each of a number of workers for each survey period (e.g., each month) as content to be presented to the viewer.

The presentation content generation unit 26 also generates, for example, a list of a worker's engagement for each survey period as content to be presented to the viewer.

(10.6) Communication Processing Unit The communication processing unit 27 controls the transmission and reception of information by the communication unit 12. The communication processing unit 27 executes the transmission step by controlling the communication unit 12. The transmission step is a step of transmitting the content generated in the presentation content generation step to a terminal. The terminal is, for example, a PC 8. The terminal has a display that displays the received content. The viewer views the content generated in the presentation content generation step via the display of the terminal.

As an example, in the transmission step, the content generated in the presentation content generation step is transmitted to the terminal at regular intervals. The regular intervals are, for example, one week, two weeks, one month, or two months.

(11) Specific Example of Engagement Estimation A specific example of engagement estimation by the engagement estimation system 1 will be described below with reference to FIG.

Worker information, including biometric information, location information, and relationship information, is collected periodically or irregularly through measurements and questionnaires given to workers. The worker information is then compiled on a monthly basis. For example, in Figure 3, worker information for January, worker information for February, etc. are compiled.

The information processing server 7 also estimates engagement based on the above-mentioned first response included in the declaration information. Here, a survey is conducted on workers once a month, and a first response is obtained as a response to the survey. Then, the information processing server 7 estimates engagement for each month based on the first response. For example, in FIG. 3, the information processing server 7 estimates engagement for January based on the first response in January, estimates engagement for February based on the first response in February, and estimates engagement for March based on the first response in March.

Once a certain amount of engagement data estimated by the information processing server 7 has been accumulated, it becomes possible for the engagement estimation system 1 to estimate engagement. In FIG. 3, after three months' worth of engagement data estimated by the information processing server 7 has been accumulated, the engagement estimation system 1 starts estimating engagement. Hereinafter, the engagement data estimated by the information processing server 7 is referred to as "reference engagement data." Also, in the example shown in FIG. 3, the first period described above corresponds to January to March before the engagement estimation system 1 starts estimating engagement.

In order to estimate engagement, the engagement estimation system 1 first determines a multiple regression equation and feature quantities. For example, when estimating engagement for April (second period), the engagement estimation system 1 refers to worker information and reference engagement data for a time period (first period) other than April (second period). In FIG. 3, the engagement estimation system 1 refers to worker information and reference engagement data for January to March. In this way, the engagement estimation system 1 determines a multiple regression equation and feature quantities. In more detail, the engagement estimation system 1 determines a multiple regression equation using engagement for January to March (first period) as the objective variable and multiple parameters extracted from worker information for January to March (first period) as multiple explanatory variables, and determines the multiple explanatory variables when the coefficient of determination of the multiple regression equation is greater than a threshold value as multiple feature quantities.

Next, the engagement estimation system 1 estimates engagement for April (second period) from the multiple regression equation and the feature quantities. More specifically, the engagement estimation system 1 finds engagement for April (second period) by substituting the feature quantities obtained from the worker information for April (second period) into the multiple regression equation.

Next, the engagement estimation system 1 generates content to be presented to the viewer based on the determined engagement, and transmits the content to the terminal (PC 8). For example, the engagement estimation system 1 generates content to be presented once a month, and transmits the content to the terminal.

By doing the same as in April, it is possible to estimate engagement after April. For example, it is possible to find engagement in May by substituting the features obtained from worker information in May into the multiple regression equation. In this case, it is not necessary to find the multiple regression equation again; the multiple regression equation found when finding engagement in April can be used.

After April (second period), it is not necessary to conduct the survey (questions) to obtain the first response. However, even after April, questions (questions included in the second question) will be asked to have the worker select the image that best represents the worker's mood from multiple images.

If the survey to obtain a first response is not conducted after April, the burden on those responding to the survey (workers, etc.) is reduced because they do not need to take time to answer the survey. Furthermore, regardless of whether the survey to obtain a first response is conducted after April, engagement from April onwards is estimated after verifying whether there is a correlation between information other than the first response (such as worker location information) and engagement. Therefore, engagement estimation system 1 can estimate engagement more objectively.

The process of determining the multiple regression equation and feature quantities does not need to be executed every time the engagement estimation system 1 estimates engagement, but only needs to be executed the first time (i.e., when estimating engagement in April).

Alternatively, the process of determining (updating) the multiple regression equation and feature quantities may be executed every time a period longer than the engagement estimation interval (one month) by the engagement estimation system 1 elapses (e.g., every six months). In this case, too, there is an advantage in that the burden is reduced, for example, because the frequency of conducting surveys to obtain a first response can be reduced.

In order for the engagement estimation system 1 to estimate engagement for a second period (e.g., April), it is not essential for the engagement estimation system 1 to collect information from the worker's information for the second period other than the multiple feature quantities determined by the feature determination unit 23. For example, assume that the conversation time between the worker and the second-level supervisor, the worker's overtime hours, and the variability in the amount of change in the worker's mood are determined as multiple feature quantities by the feature determination unit 23. In this case, it is not essential for the engagement estimation system 1 to collect information from the worker's information for the second period other than the multiple feature quantities determined by the feature determination unit 23 (e.g., the worker's activity level).

In addition, to estimate engagement for April (second period), the engagement estimation system 1 may determine a multiple regression equation and multiple feature quantities by referring to worker information and reference engagement data for a period later than April (second period) (e.g., May to June).

(12) Specific Example of the First Step Next, a specific example of the flow of the first step will be described with reference to Fig. 7. The first step is a step of executing a questioning step at multiple points in time (morning and evening of each day) in which an operation window 820 is displayed on the display unit 82 of the PC 8 and the worker is prompted to select one of the images.

First, the PC 8 determines whether the current time is a specific time in the morning or evening (step ST1). When the specific time arrives (step ST1: Yes), the PC 8 displays a question on the display unit 82 (step ST2). More specifically, the PC 8 displays an operation window 820 on the display unit 82, which presents a question for prompting the worker to select an image that matches the worker's mood from among multiple images.

The worker answers the question by selecting one of the images. When the question is answered (step ST3: Yes), the processing unit 81 stores the answer time and the answer content in the memory unit 84 (step ST4). The answer content is, in other words, a numerical value of 1, 2, 3, or 4 as answer information. The processing unit 81 then transmits the answer time and the answer content to the engagement estimation system 1 (step ST5).

If no reply is made within a certain period of time after the operation window 820 is displayed on the display unit 82 (step ST3: No), the processing unit 81 issues a notification prompting the reply (step ST6). Until a reply is made, the processing unit 81 issues a notification prompting the reply at regular intervals.

The series of processes shown in Figure 7 is performed every day, and response information for the morning and evening of each day is obtained by the engagement estimation system 1.

Note that the flowchart shown in FIG. 7 merely shows one example of the flow of processing performed by PC 8, and the order of processing may be changed as appropriate, and processing may be added or omitted as appropriate.

(13) Details of Feature A feature may be, for example, related to "job resources,""personalresources," or "job demands" defined in the "job demands-resources model," i.e., the "JD-R model." In addition, the feature may be related to empathy or a sense of satisfaction with at least one of the vision, mission, and philosophy of the group (company, etc.) to which the worker belongs for work. Alternatively, the feature may be related to two or more of the above.

As an example, "job resources" refer to at least one of "support from those around you," "relationships with those around you," "job autonomy," "coaching from colleagues," "feedback from colleagues," "diversity of relationships," and "opportunities for career development."

As an example, "personal resources" may relate to at least one of "optimism," "resilience," and "recovery status."

As an example, "job demands" refers to at least one of the following: "quantitative workload," "qualitative workload," and "physical workload at work."

As described above, the feature determination unit 23 extracts multiple parameters from, for example, the worker's information and determines multiple feature amounts from them. The parameters (feature amounts) are, for example, the conversation time between the worker and a specific person, the worker's activity level, and the worker's overtime hours. The multiple feature amounts determined by the feature determination unit 23 include at least one of the following: a feature amount based on the worker's biometric information measured by the biometric information measurement terminal 3, a feature amount based on the worker's location information at the worker's workplace, a feature amount based on relationship information regarding the relationship between the worker and another worker at the workplace, and a feature amount based on the usage history of the computer system (PC 8) used by the worker for work. The multiple feature amounts determined by the feature determination unit 23 further include a feature amount based on response information.

Below are examples of feature quantities. However, feature quantities are not limited to those listed below.

(13.1) Work Resources The conversation time between a worker and a specific person as a feature is related to the "work resources" of "support from the surroundings" and "relationships with the surroundings". The (face-to-face) conversation time between a worker and a specific person can be obtained, for example, from the position information and voice information of each of a plurality of workers. In other words, the conversation time between a worker and a specific person can be determined as the time when the distance between the worker and the specific person is within a predetermined distance (for example, 1 meter) and voice information is output from a microphone present around the conversation location. The microphone may be, for example, carried by the worker or installed near the worker's work location.

Furthermore, the duration of a (face-to-face) conversation between a worker and a specific person can also be calculated, for example, from only the location information of each of a number of workers. In other words, the duration of a conversation can be calculated by conveniently determining that the worker and a specific person are in close proximity to each other as being for the purpose of conversation. For example, the duration of a conversation between a worker and a specific person can be determined as the time when the distance between them is within a specified distance.

Furthermore, the conversation time between the worker and a specific person is not limited to face-to-face conversation time, but may also include non-face-to-face (e.g., online) conversation time. Online conversation time is extracted, for example, from the usage history of PC8.

　Furthermore, the amount of time a worker has face-to-face conversations with a specific person and the amount of time he or she has not face-to-face conversations may be required separately.

Furthermore, the conversation time between the worker and the specific person may be extracted from the declaration information input to the operation terminal 6. In other words, the engagement estimation system 1 may obtain the conversation time between the worker and the specific person based on the declaration from the respondent (e.g., the worker).

As described above, the relationship information is information about the relationships between multiple workers in the workplace of the multiple workers. The feature determination unit 23 may obtain the conversation time of multiple workers for each relationship based on the relationship information. That is, the feature determination unit 23 may obtain the conversation time between a worker and a worker in a specific position. For example, the feature determination unit 23 may obtain the conversation time between a worker and a superior, the conversation time between a worker and a subordinate, and the conversation time between workers of the same position. The relationships may also be further subdivided. For example, the feature determination unit 23 may obtain the conversation time between a worker and a first-level superior (immediate superior) and the conversation time between a worker and a second-level superior (the superior of the immediate superior). For example, the feature determination unit 23 may obtain the conversation time between workers who belong to the same organization (department, etc.) and between workers who belong to different organizations (departments, etc.). Also, for example, the feature determination unit 23 may determine the conversation time between workers who are in charge of the same work (project, etc.).

Furthermore, the number of conversations between a worker and a specific person may be obtained as a feature. The number of face-to-face conversations can be obtained, for example, from the position information of each of a plurality of workers. In other words, the number of times the distance between the worker and a specific person changes from more than a specified distance (for example, one meter) to within the specified distance can be regarded as the number of conversations between the worker and the specific person.

The feature quantities, which are the worker's overtime hours, the number of overtime hours, the number of night shift hours, and the number of holiday shift hours, are related to the "work discretion" of the "work resources." The worker's overtime hours, the number of overtime hours, the number of night shift hours, and the number of holiday shift hours are extracted, for example, from the attendance information output from the attendance management system 9, or the reported information entered into the operation terminal 6. Alternatively, the worker's overtime hours, the number of overtime hours, the number of night shift hours, and the number of holiday shift hours are determined, for example, from the start and end times of PC 8, which are extracted from the usage history of PC 8.

The mood at the end of the workday, as a feature, is related to the "work resources" of "coaching from colleagues" and "feedback from colleagues." The mood at the end of the workday is extracted, for example, from the reporting information entered into the operation terminal 6.

The number of communications within a department and the number of people with whom communication takes place within the department, as features, are related to the "diversity of human relationships" of "work resources." The number of face-to-face communications within a department and the number of people with whom communication takes place within the department can be determined, for example, from the location information of each of multiple workers, or from location information and voice information, or can be extracted from reporting information entered into the operation terminal 6. The number of non-face-to-face (e.g., online) communications within a department and the number of people with whom communication takes place within the department can be extracted, for example, from the usage history of PC 8 or from reporting information entered into the operation terminal 6.

The number of spaces used and the number of times specialized tools were used, as features, are related to the "opportunities for career development" of the "work resources." The number of spaces used and the number of times specialized tools were used can be obtained, for example, from the location information of the worker, or extracted from the usage history of the PC 8 or the reporting information entered into the operation terminal 6.

(13.2) Personal Resources The variation in mood change between morning and evening as a feature quantity is related to “optimism” of “personal resources.” The variation in mood change between morning and evening is calculated based on the answers entered into the PC 8.

The feature quantities of break time, time when no data was entered into the PC, and the number of times when no data was entered into the PC (the number of times when no data was entered into the PC for a specified period of time or more) are related to the "resilience" of "individual resources." The break time, time when no data was entered into the PC, and the number of times when no data was entered into the PC are extracted, for example, from the usage history of the PC 8 or the reported information entered into the operation terminal 6.

The amount of movement within the office, as a feature, is related to the "resilience" of "personal resources." The amount of movement within the office can be calculated, for example, from the location information of the worker, or measured by a pedometer (movement measuring terminal 10) carried by the worker.

The PC operation time and the number of times the PC is operated before the start of work, after the end of work, during late night hours, and on holidays, as feature quantities, relate to the "recovery status" of "personal resources." The PC operation time and the number of times the PC is operated before the start of work, after the end of work, during late night hours, and on holidays are extracted, for example, from the usage history of the PC 8, the attendance information output from the attendance management system 9, or the reporting information input to the operation terminal 6.

The work time interval (the length of time between the end of work on a certain day and the start of work on the following day) as a feature quantity is related to the "recovery status" of the "personal resources." The work time interval is extracted, for example, from the usage history of the PC 8, the attendance information output from the attendance management system 9, or the reporting information input to the operation terminal 6.

The volume of the worker's voice, as a feature, is related to the "recovery status" of the "personal resources." The volume of the worker's voice is measured, for example, by a microphone provided in the bio-information measurement terminal 3.

(13.3) Work Demands The feature quantity, ie, the working time (the length of time from the time of entering the workplace to the time of leaving the workplace), is related to the "quantitative burden of work" of the "work demands." The working time can be obtained, for example, from the location information of the worker or the attendance information output from the attendance management system 9.

The amount of time spent using a PC after work hours, as a feature, is related to the "quantitative burden of work" of the "demands of work." The amount of time spent using a PC after work hours is extracted, for example, from the usage history of PC8.

The feature quantities, ie, the time spent at a rest area and the number of times the rest area is used, relate to the "quantitative workload" of the "job demands." The time spent at a rest area and the number of times the rest area is used can be determined, for example, from a combination of the worker's location information and area information related to the rest area, etc. The area information is obtained, for example, from the data server 5.

The feature quantities, ie, the time period during which there is no PC input between the start and end of work and the number of times (the number of times that no input continues for a specified period of time or more), relate to the "quantitative workload" of the "demand level of work." The time period during which there is no PC input between the start and end of work and the number of times, are extracted, for example, from the usage history of PC8.

The amount of conversation in the workplace, as a feature, is related to the "demands of work" and the "quantitative burden of work." The amount of conversation in the workplace can be obtained, for example, from the output (audio information) of the microphone equipped in the bioinformation measurement terminal 3.

The number of keyboard operations per unit time and the amount of mouse cursor movement per unit time, which are characteristic quantities, are related to the "quality of work burden" of the "job demands." The number of keyboard operations per unit time and the amount of mouse cursor movement per unit time are extracted, for example, from the usage history of the PC 8.

The PC operation time, as a feature, is related to the "quality of work burden" of the "job demands." The PC operation time is extracted, for example, from the usage history of PC8.

The task time with a high qualitative burden, as a feature, is related to the "qualitative burden of work" of the "degree of work demands." The task time with a high qualitative burden can be determined, for example, from bioinformation measured by the bioinformation measurement terminal 3. As a specific example, the engagement estimation system 1 assumes that a state in which the heart rate, as bioinformation, is higher than a corresponding threshold value is a state of high qualitative burden, and determines the accumulated time in the state of high qualitative burden as the task time with a high qualitative burden.

The amount of activity of a worker, as a feature, is related to the "physical burden of work" of the "job demands." The amount of activity of a worker is expressed, for example, in METs. The amount of activity of a worker can be calculated, for example, from bio-information (heart rate, blood pressure, skin temperature, amount of sweat, etc.) measured by the bio-information measuring terminal 3.

The number of steps taken by a worker, as a feature, is related to the "physical burden of work" of the "job demands." The number of steps taken by a worker can be determined, for example, from the worker's location information. Alternatively, the number of steps taken by a worker can be measured, for example, by a pedometer (exercise measuring terminal 10) carried by the worker.

The maximum heart rate of a worker, as a feature, is related to the "physical burden of work" of the "job demands." The maximum heart rate of a worker is extracted, for example, from heart rate measurement data as biometric information.

(13.4) Degree of Sympathy with Ideals A feature related to the degree of sympathy with ideals (sympathy or a sense of agreement with at least one of the vision, mission, and ideals of the group to which the worker belongs for work) is extracted, for example, from the declaration information input to the operation terminal 6. The feature related to the degree of sympathy with ideals is, for example, an index indicating the extent to which the worker understands the meaning of the work.

(14) Modifications Several modifications of the embodiment will be described below. Hereinafter, the above-described aspects of the embodiment will be referred to as basic examples.

(14.1) Modification 1
In this modification, the feature based on the response information includes a change amount of the worker's mood between multiple points in time, i.e., one of the multiple feature amounts is the change amount.

The worker answers the questions during the first period, and the amount of change during the second period is obtained. Also, during the first period, the remaining feature amounts among the multiple feature amounts during the first period are obtained by the bioinformation measuring terminal 3, the position measuring system 4, the data server 5, etc.

The worker answers the questions during the second time period, thereby obtaining the amount of change during the second time period. Also, during the second time period, the remaining feature quantities among the multiple feature quantities during the second time period are obtained in the bioinformation measurement terminal 3, the position measurement system 4, the data server 5, etc.

The regression unit 24 obtains a regression equation that represents the relationship between the multiple feature amounts and engagement in the first time period. The estimation unit 25 estimates the worker's engagement in the second time period based on the multiple feature amounts in the second time period and the regression equation.

As an example, if a regression equation is obtained for a first worker that shows the correlation between the amount of change in mood and engagement, the greater the amount of change, the lower the engagement. Note that if a regression equation is obtained for a second worker different from the first worker that shows the correlation between the amount of change and engagement, the ratio of change in engagement to the amount of change may be different compared to the first worker, or the greater the amount of change, the higher the engagement.

In the basic example, the second period is one month (see FIG. 3). In contrast, in this modified example, the second period may be, for example, one day. That is, the amount of change in mood from morning to evening on a certain day may be used as a feature to estimate engagement on the same day. The second period may also be a period shorter than one day (for example, a few hours).

For example, if the second period includes June 5th and June 6th, the amount of change in mood from morning to evening on June 5th and the amount of change in mood from morning to evening on June 6th can be obtained. In this way, when multiple amounts of change are obtained in the second period, the average value of the multiple amounts of change can be used as the feature amount.

(14.2) Modification 2
In this modification, the feature quantity based on the response information is a parameter corresponding to the worker's mood at each of a plurality of time points. In other words, the

numerical values

1, 2, 3, or 4 as the response information in the morning and evening, respectively, are themselves feature quantities based on the response information.

In addition, if multiple pieces of answer information are obtained in the second period, the average value of the multiple pieces of answer information may be used as the feature amount. For example, if the second period is from the morning of June 5th to the evening of June 6th, in the example shown in FIG. 5, the average value of the four pieces of answer information, (2+2+3+2)/4=2.25, may be used as the feature amount.

As an example, when a regression equation showing the correlation between response information and engagement is obtained for a first worker, the larger the numerical value showing the response information, the lower the engagement. Note that when a regression equation showing the correlation between the numerical value showing the response information and engagement is obtained for a second worker different from the first worker, the rate of change in engagement relative to the change in the numerical value showing the response information may be different compared to the case of the first worker, and, for example, the closer the numerical value showing the response information is to a specified numerical value between 1 and 4 (e.g., 2), the higher the engagement may be.

In the basic example, the second period is one month (see FIG. 3). In contrast, in this modified example, the second period may be, for example, one day, or may be a period shorter than one day (for example, a few hours).

(14.3) Other Modifications of the Embodiments Other modifications of the embodiment are listed below. The following modifications may be realized in appropriate combination. The following modifications may also be realized in appropriate combination with the above-mentioned

modification

1 or 2.

The engagement estimation system 1 may be equipped with a display device that displays the information generated in the presentation content generation step.

The engagement estimation system 1 may include an operation unit that accepts operations for generating the declaration information, and may also serve as the operation terminal 6.

The engagement estimation system 1 may include a PC 8 or the like as a worker terminal that performs the above-mentioned questioning steps at multiple points in time.

In a basic example, the interrogation step is performed twice per day. Alternatively, the interrogation step may be performed once per day, or three or more times per day. Alternatively, the interrogation step may be performed, for example, every other day or every few days.

In a basic example, the amount of change in the worker's mood between multiple points in time is the amount of change in the worker's mood from a point in time within a day to a point in time on the same day. In contrast, the amount of change in the worker's mood between multiple points in time may be the amount of change in the worker's mood from one day to another day.

Each of the multiple images showing human facial expressions displayed in the operation window 820 is not limited to illustrations and may be, for example, photographs.

The moods represented by the multiple images displayed in the operation window 820 are not limited to very good, good, bad, or very bad. The moods may be, for example, a relaxed mood, a tense mood, or a tired mood.

Each of the multiple images displayed in the operation window 820 is not limited to images showing a person's facial expression, and may be an image or symbol showing a person's mood. The image or symbol showing a person's mood may be, for example, an image or symbol of sweat showing a tired mood, an image or symbol such as an eighth note or a heart mark showing a good mood, an image or symbol of an arrow pointing up to the right showing a good mood, or an image or symbol of an arrow pointing down to the right showing a depressed mood.

Furthermore, the multiple images displayed in the operation window 820 may be identical to each other, and two or more of the multiple images may be selectable. For example, the multiple images displayed in the operation window 820 may be multiple star marks. The more star marks selected, the better the mood.

The multiple images displayed in the operation window 820 may also be multiple weather marks. The multiple weather marks may include, for example, sunny, cloudy, and rainy marks. The sunny mark represents a better mood than the cloudy mark. The cloudy mark represents a better mood than the rainy mark. The multiple weather marks may also include, for example, a mark that combines a sunny mark and a cloudy mark. This mark represents a worse mood than the sunny mark, but a better mood than the cloudy mark.

In a basic example, four images representing four moods of a worker are displayed on the display unit 82 of the PC 8. However, the number of images displayed (i.e., the number of moods) may be two, three, five or more.

In the basic example, the mood represented by each image is displayed in text near (below) each image in the operation window 820. However, it is not essential that such text be displayed in the operation window 820.

The PC 8 may execute a process of displaying a question for prompting the worker to select an image that matches the worker's mood from among a plurality of images, based on control by the engagement estimation system 1. For example, the engagement estimation system 1 may transmit a predetermined command signal to the PC 8 at a predetermined time period, and the PC 8 may display a question on the display unit 82 upon receiving the command signal.

The feature based on the response information is the variability in the amount of change in the worker's mood in the basic example, the amount of change in the worker's mood in variant 1, and the response information itself in variant 2. Two or three of these may be used as feature based on the response information to estimate engagement.

The feature determination unit 23 may refer to a correlation coefficient instead of the coefficient of determination to determine the strength of correlation between multiple parameters and engagement in the first period.

The entity that executes the engagement estimation system 1 or the engagement estimation method in the present disclosure includes a computer system. The computer system is mainly composed of a processor and a memory as hardware. At least a part of the functions of the entity that executes the engagement estimation system 1 or the engagement estimation method in the present disclosure is realized by the processor executing a program recorded in the memory of the computer system. The program may be pre-recorded in the memory of the computer system, may be provided through a telecommunications line, or may be provided by being recorded on a non-transitory recording medium such as a memory card, an optical disk, or a hard disk drive that is readable by the computer system. The processor of the computer system is composed of one or more electronic circuits including a semiconductor integrated circuit (IC) or a large scale integrated circuit (LSI). The integrated circuits such as ICs or LSIs referred to here are called different names depending on the degree of integration, and include integrated circuits called system LSIs, VLSIs (Very Large Scale Integration), or ULSIs (Ultra Large Scale Integration). Furthermore, a field-programmable gate array (FPGA) that is programmed after the LSI is manufactured, or a logic device that allows the reconfiguration of the connection relationships within the LSI or the reconfiguration of the circuit partitions within the LSI, can also be used as a processor. Multiple electronic circuits may be integrated into one chip, or may be distributed across multiple chips. Multiple chips may be integrated into one device, or may be distributed across multiple devices. The computer system referred to here includes a microcontroller having one or more processors and one or more memories. Thus, the microcontroller is also composed of one or more electronic circuits including a semiconductor integrated circuit or a large-scale integrated circuit.

Furthermore, it is not essential for the engagement estimation system 1 that multiple functions are consolidated into one device, and multiple components of the engagement estimation system 1 may be distributed across multiple devices. Furthermore, at least some of the functions of the engagement estimation system 1 may be realized by a server or a cloud (cloud computing), etc.

On the other hand, in the basic example, multiple functions that are distributed across multiple devices may be consolidated into one device. For example, at least two of the data server 5, the information processing server 7, and the engagement estimation system 1 may be consolidated into one device. Also, for example, the PC 8 may also function as the operation terminal 6. Also, for example, the bio-information measuring terminal 3 may also function as the exercise measuring terminal 10.

At least a portion of the functions of the engagement estimation system 1 may be realized by a computational model generated by machine learning. For example, the third step of determining a plurality of feature quantities may be realized by the computational model.

(summary)
The above-described embodiments and the like disclose the following aspects.

The engagement estimation method according to the first aspect is a method for estimating a worker's engagement with work, which is executed by an engagement estimation system (1). The engagement estimation method has a first step, a second step, a third step, a fourth step, and a fifth step. In the first step, a questioning step is carried out at multiple time points. In the questioning step, multiple images corresponding one-to-one to multiple moods are displayed, and the worker is asked to select an image that corresponds to the worker's mood from the multiple images. In the second step, answer information regarding which image was selected from the multiple images at each of the multiple time points is stored. In the third step, multiple feature amounts are determined. In the fourth step, a regression equation is obtained that represents the relationship between the multiple feature amounts in the first period and the worker's engagement in the first period that is determined in advance. In the fifth step, the worker's engagement in the second period is estimated based on the multiple feature amounts in a second period different from the first period and the regression equation. The multiple feature amounts include at least one of the following: a feature amount based on the worker's biometric information measured by the biometric information measuring terminal (3), a feature amount based on the worker's location information at the worker's workplace, a feature amount based on relationship information regarding the relationship between the worker and another worker at the workplace, and a feature amount based on the usage history of a computer system (PC8) used by the worker for work. The multiple feature amounts further include a feature amount based on response information.

In the above configuration, the worker selects an image that matches the worker's mood from among multiple images, and the engagement estimation system (1) estimates the worker's engagement. The worker can intuitively select an image that matches his or her current mood. This reduces the burden on the worker in answering questions. For example, the burden on the worker is reduced compared to when a question is displayed, the worker reads the question, and then the worker answers the question.

In addition, in the engagement estimation method according to the second aspect, in the first aspect, in the fourth step, a regression equation is calculated based on a plurality of feature quantities in the first period and the engagement of the worker in the first period that has been calculated in advance.

With the above configuration, the process of calculating the regression equation and the process of estimating worker engagement in the second time period can be consolidated into the engagement estimation system (1).

In addition, the engagement estimation method according to the third aspect, in the first or second aspect, further includes a presentation content generation step of generating content to be presented to the viewer based on the estimation result in the fifth step.

The above configuration allows the viewer to know the estimation results.

In addition, the engagement estimation method according to the fourth aspect, in the third aspect, further includes a transmission step of transmitting the content generated in the presentation content generation step to the terminal.

The above configuration allows the viewer to know the estimation results.

In addition, in the engagement estimation method according to the fifth aspect, in any one of the first to fourth aspects, in the first step, the questioning step is carried out at multiple time periods in one day.

With the above configuration, the response information includes information regarding changes in the worker's mood throughout the day. This allows the engagement estimation system (1) to effectively estimate engagement.

In addition, in the engagement estimation method according to the sixth aspect, in any one of the first to fifth aspects, the third step includes a step of determining the amount of change in the worker's mood between multiple points in time. The feature based on the response information includes the variance in the amount of change.

　With the above configuration, the engagement estimation system (1) can effectively estimate engagement.

In addition, in the engagement estimation method according to the seventh aspect, in any one of the first to sixth aspects, the third step includes a step of determining the amount of change in the worker's mood between multiple points in time. The feature based on the response information includes the amount of change.

In addition, in the engagement estimation method according to the eighth aspect, in any one of the first to seventh aspects, the feature based on the response information is a parameter corresponding to the worker's mood at each of a plurality of points in time.

In addition, the engagement estimation method according to the ninth aspect, in any one of the first to eighth aspects, further includes a notification step of issuing a notification to prompt a selection if an image that matches the worker's mood is not selected in the first step.

The above configuration reduces the possibility that a worker will forget to respond, which makes it impossible for the engagement estimation system (1) to estimate their engagement.

In addition, in the engagement estimation method according to the tenth aspect, in any one of the first to ninth aspects, each of the multiple images displayed in the first step is an image showing a person's facial expression.

The above configuration makes it easy for workers to intuitively select an image that matches their mood.

Configurations other than the first aspect are not essential to the engagement estimation method and may be omitted as appropriate.

The program according to the eleventh aspect is a program for causing one or more processors of a computer system (of the engagement estimation system 1) to execute the engagement estimation method according to any one of the first to tenth aspects.

The above configuration reduces the burden on workers when responding.

Furthermore, an engagement estimation system (1) relating to a twelfth aspect estimates a worker's engagement with work. The engagement estimation system (1) includes an acquisition unit (21), a memory unit (11), a feature determination unit (23), a regression unit (24), and an estimation unit (25). The acquisition unit (21) acquires answer information from a worker terminal (PC8). The worker terminal performs a questioning step at multiple time points. In the questioning step, multiple images corresponding one-to-one to multiple moods are displayed, and the worker is prompted to select an image from the multiple images that matches the worker's mood. The answer information is information regarding which image was selected from the multiple images at each of the multiple time points. The memory unit (11) stores the answer information. The feature determination unit (23) determines multiple feature amounts. The regression unit (24) acquires a regression equation expressing the relationship between the multiple feature amounts in the first period and the worker's engagement in the first period, which has been determined in advance. The estimation unit (25) estimates the worker's engagement in the second period based on the multiple feature amounts in the second period different from the first period and the regression equation. The multiple feature amounts include at least one of a feature amount based on the worker's biometric information measured by the biometric information measurement terminal (3), a feature amount based on the worker's position information in the worker's workplace, a feature amount based on relationship information regarding the relationship between the worker and another worker in the workplace, and a feature amount based on the usage history of a computer system (PC8) used by the worker for work. The multiple feature amounts further include a feature amount based on response information.

The above configuration reduces the burden on workers when responding.

In addition, in the engagement estimation system (1) according to the thirteenth aspect, in the twelfth aspect, the feature determination unit (23) determines the amount of change in the worker's mood between multiple points in time. The feature based on the response information includes the variability in the amount of change.

Configurations other than the twelfth aspect are not essential to the engagement estimation system (1) and may be omitted as appropriate.

Not limited to the above aspects, various configurations (including modified examples) of the engagement estimation system (1) according to the embodiment can be embodied in an engagement estimation method, a (computer) program, or a non-transitory recording medium having a program recorded thereon.

1 Engagement estimation system 3 Biometric information measurement terminal 8 PC (worker terminal, computer system)
11 Storage unit 21 Acquisition unit 23 Feature amount determination unit 24 Regression unit 25 Estimation unit

Claims

An engagement estimation method for estimating a worker's engagement in a job, the method being executed by an engagement estimation system, the method comprising:
A first step of displaying a plurality of images corresponding one-to-one to a plurality of moods and asking the worker to select an image that corresponds to the worker's mood from the plurality of images at a plurality of time points;
a second step of storing answer information regarding which image of the plurality of images was selected at each of the plurality of time points;
a third step of determining a plurality of features;
A fourth step of acquiring a regression equation expressing a relationship between the plurality of feature amounts in a first period and the engagement of the worker in the first period, the relationship being determined in advance;
and a fifth step of estimating the engagement of the worker during a second period based on the plurality of feature amounts during the second period and the regression equation, the fifth step comprising:
The plurality of feature amounts are
A feature based on the worker's biometric information measured by a biometric information measuring terminal;
A feature based on location information of the worker at the workplace of the worker;
A feature based on relationship information regarding a relationship between the worker and another worker in the workplace;
and a feature based on a usage history of a computer system used by the worker for the job;
the plurality of feature amounts further includes a feature amount based on the answer information,
Engagement Estimation Methodology.
In the fourth step, the regression equation is obtained based on the plurality of feature amounts in the first period and the engagement of the worker in the first period that is obtained in advance.
The method for estimating engagement according to claim 1 .
The method further includes a presentation content generating step of generating content to be presented to a viewer based on the estimation result in the fifth step.
The method for estimating engagement according to claim 1 .
The method further includes a transmission step of transmitting the content generated in the presentation content generation step to a terminal.
The method for estimating engagement according to claim 3 .
In the first step, the questioning step is carried out at a plurality of time periods in a day.
The method for estimating engagement according to claim 1 .
The third step includes a step of determining a change in the mood of the worker between the plurality of time points,
the feature based on the response information includes a variance in the amount of change;
The method for estimating engagement according to claim 1 .
The third step includes a step of determining a change in the mood of the worker between the plurality of time points,
the feature based on the answer information includes the amount of change;
The method for estimating engagement according to claim 1 .
The feature based on the response information is a parameter corresponding to the mood of the worker at each of the plurality of time points.
The method for estimating engagement according to claim 1 .
The method further includes a notification step of issuing a notification to prompt a selection when the image corresponding to the mood of the worker is not selected in the first step.
The method for estimating engagement according to claim 1 .
Each of the plurality of images displayed in the first step is an image showing a human facial expression.
The method for estimating engagement according to claim 1 .
2. A method for estimating engagement according to claim 1, comprising:
program.
An engagement estimation system for estimating a worker's engagement with work, comprising:
an acquisition unit that acquires answer information from a worker terminal that performs a questioning step at multiple time points, the questioning step displaying multiple images corresponding one-to-one to multiple moods and having the worker select an image that corresponds to the worker's mood from the multiple images, the acquisition unit acquiring answer information regarding which image was selected from the multiple images at each of the multiple time points;
A storage unit that stores the answer information;
a feature determining unit for determining a plurality of feature amounts;
A regression unit that acquires a regression equation that expresses a relationship between the plurality of feature amounts in a first time period and the engagement of the worker in the first time period that is obtained in advance;
an estimation unit that estimates the engagement of the worker during a second period based on the plurality of feature amounts during the second period and the regression equation,
The plurality of feature amounts are
A feature based on the worker's biometric information measured by a biometric information measuring terminal;
A feature based on location information of the worker at the workplace of the worker;
A feature based on relationship information regarding a relationship between the worker and another worker in the workplace;
and a feature based on a usage history of a computer system used by the worker for the job;
the plurality of feature amounts further includes a feature amount based on the answer information,
Engagement Estimation System.
The feature determining unit determines an amount of change in the mood of the worker between the plurality of time points;
the feature based on the response information includes a variance in the amount of change;
The engagement estimation system of claim 12 .