Information processing device, information processing system, churn probability analysis method, and control program

Abstract

To provide information useful for grasping in advance a user's discomfort about devices.SOLUTION: An information processing apparatus 10 comprises: an acquisition unit 111 that acquires operating log data of devices present in a nursing and care facility 900; an analysis unit 112 that analyzes the possibility of withdrawal of a user who uses the devices in the nursing and care facility 900 based on the operating log data; and an output unit 113 that outputs a result of the analysis performed by the analysis unit 112.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing system, a churn probability analysis method, and a control program. [Background technology]

[0002] In nursing care facilities such as nursing homes for the elderly and hospitals, residents or hospitalized patients (hereinafter simply referred to as residents) may fall while walking or fall out of bed and injure themselves in the nursing care facility. Therefore, to enable nurses or caregivers to rush to the scene immediately when a resident falls into such a state, development is underway for a system that constantly monitors the condition of the resident (for example, Patent Document 1).

[0003] A monitoring support system that aims to reduce the burden on users of such systems is known (for example, Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-90913 [Patent Document 2] Japanese Patent Application Publication No. 2019-21002 Summary of the Invention [Problem to be solved by the invention]

[0005] Although Patent Document 2 is a system that reduces the burden on users, users such as nursing staff have the impression that ICT and IoT devices are difficult to operate and that using them increases the burden on them, making it difficult to understand the dissatisfaction of device users, and there was a problem that simply introducing the system's devices did not lead to a reduction in the burden on device users. Such problems led to the devices being discontinued, and in the past, this was addressed by receiving complaints from device users, but it was only possible to understand the dissatisfaction of device users after a complaint was received, making it impossible to respond early.

[0006] The present invention has been made to solve such problems, and aims to provide information that is useful for understanding the dissatisfaction of device users in advance. [Means for solving the problem]

[0007] The above object of the present invention can be achieved by the following means.

[0008] (1) an acquisition unit that acquires operation log data of devices present in a nursing care facility; an analysis unit that analyzes the likelihood of a user using the device leaving the device based on the operation log data; an output unit that outputs the analysis result by the analysis unit; An information processing device comprising:

[0009] (2) The acquisition unit acquires staff response information log data through the staff terminal, The information processing device according to (1), wherein the analysis unit analyzes the likelihood of withdrawal based on the operation log data and the corresponding information log data.

[0010] (3) The information processing device according to (2) above, wherein the analysis unit calculates a plurality of indicators of the likelihood of churn by analyzing the operation log data and the corresponding information log data for a predetermined period of time.

[0011] (4) The indicator of the likelihood of withdrawal includes an indicator that evaluates the operation or utilization of the device, The information processing device described in (3) above, wherein the analysis unit determines, for each of the indicators, whether the indicator is outside a predetermined range or whether the rate of change of the indicator per predetermined period is outside a predetermined range, and analyzes the possibility of withdrawal based on a stage determined according to the number of indicators determined to be outside the predetermined range.

[0012] (5) The information processing device according to any one of (1) to (3), wherein the analysis unit analyzes the likelihood of withdrawal using a trained model.

[0013] (6) The information processing device according to any one of (1) to (5), wherein the output unit outputs a response action for the dropout in the analysis result.

[0014] (7) The information processing device according to (6), wherein the response actions are plural, and the output unit includes the priority of the response actions in the output.

[0015] (8) The acquisition unit acquires the operation log data from each of the plurality of nursing care facilities, The analysis unit analyzes the possibility of withdrawal in each of the plurality of nursing care facilities, The information processing device described in any one of (1) to (7) above, wherein the output unit outputs the analysis results for each of the multiple nursing care facilities for each nursing care facility.

[0016] (9) a display device having a display unit that displays the analysis results; An information processing device according to any one of (1) to (8) above; An information processing system comprising:

[0017] (10) Furthermore, the equipment present in the nursing care facility, The information processing system according to (9) above.

[0018] (11) Step (a) of acquiring operation log data of devices present in a nursing care facility; (b) analyzing the likelihood of users using the device dropping out based on the operation log data; Step (c) of outputting the analysis results of step (b); A method for analyzing the likelihood of churn, comprising:

[0019] (12) In the step (a), log data of staff response information is acquired through the staff terminal; The churn possibility analysis method according to (11) above, wherein in the step (b), the churn possibility is analyzed based on the operation log data and the response information log data.

[0020] (13) The churn probability analysis method according to (12) above, wherein in step (b), a plurality of churn probability indicators are calculated by analyzing the operation log data and the response information log data for a predetermined period.

[0021] (14) The indicator of the likelihood of withdrawal includes an indicator evaluating the operation or utilization of the device, In the step (b), for each of the indicators, it is determined whether the indicator is outside a predetermined range or whether the rate of change of the indicator per predetermined period is outside a predetermined range, and the churn possibility is analyzed based on the stage determined according to the number of indicators determined to be outside the predetermined range.

[0022] (15) A method for analyzing the likelihood of churn described in any one of (11) to (13) above, wherein in step (b), the likelihood of churn is analyzed using a trained model.

[0023] (16) The churn likelihood analysis method according to any one of (11) to (15) above, wherein in step (c), a corresponding action is output in response to churn in the analysis result.

[0024] (17) The method for analyzing the likelihood of withdrawal according to (16) above, wherein the response actions are plural, and in step (c), the priority of the response actions is included in the output.

[0025] (18) In the step (a), the operation log data is acquired from each of the plurality of nursing care facilities; In the step (b), the possibility of withdrawal is analyzed for each of the plurality of nursing care facilities; A method for analyzing the likelihood of withdrawal described in any one of (11) to (17) above, wherein in step (c), the analysis results for each of the multiple nursing care facilities are output for each nursing care facility.

[0026] (19) A control program for causing a computer to execute the churn likelihood analysis method described in any one of (11) to (18) above. [Effects of the Invention]

[0027] The information processing device according to the present invention includes an acquisition unit that acquires operation log data of devices present in a nursing care facility, an analysis unit that analyzes the likelihood of users of the devices dropping out based on the operation log data, and an output unit that outputs the analysis results obtained by the analysis unit. This makes it possible to provide information that is useful for identifying device user dissatisfaction in advance. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a schematic diagram showing a plurality of nursing care facilities in which an information processing system according to this embodiment is installed, and a service provider. [Figure 2] 1 is a block diagram showing a schematic configuration of an information processing apparatus according to a first embodiment. [Figure 3A] 10 is a table showing an example of operation log data. [Figure 3B] 10 is a table showing an example of response information log data. [Figure 4] FIG. 2 is a block diagram showing a schematic configuration of a terminal device. [Figure 5]1 is a schematic diagram showing the overall configuration of a monitoring support system used in nursing care facilities. [Figure 6] FIG. 10 is a diagram illustrating an example of a detection unit installed in a resident's room. [Figure 7] FIG. 2 is a block diagram showing a schematic configuration of a detection unit. [Figure 8] FIG. 1 is a block diagram showing a schematic configuration of a monitoring device. [Figure 9] FIG. 2 is a block diagram showing a schematic configuration of a terminal device. [Figure 10] FIG. 2 is a block diagram showing a schematic configuration of a staff terminal. [Figure 11] 10 is a flowchart showing an event response process in the entire watching support system. [Figure 12] 10 is an example of an event list recorded in a storage unit. [Figure 13] 10 is an example of an operation screen for responding to an event notification. [Figure 14] 10 is an example of an operation screen for setting notifications. [Figure 15] FIG. 10 is a diagram showing notification settings and types of notification settings. [Figure 16] 10 is a flowchart showing a withdrawal possibility analysis process. [Figure 17] 10 is a subroutine flowchart showing the processing of step S44. [Figure 18] 1 is a table showing an example of a churn likelihood index. [Figure 19] This is a table showing the indicators classified into factors 1 and 2 used to determine the stage. [Figure 20] 10 is a table showing indexes classified by phase used to determine the stage. [Figure 21] FIG. 10 is a schematic diagram illustrating stages used to determine the possibility of departure. [Figure 22] FIG. 10 is a block diagram showing a schematic configuration of an information processing device according to a second embodiment. [Figure 23] 10 is a subroutine flowchart showing the processing of step S44. [Figure 24]1 is a flowchart showing a machine learning method for a trained model. DETAILED DESCRIPTION OF THE INVENTION

[0029] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. In the description of the drawings, the same elements are designated by the same reference numerals, and duplicate explanations will be omitted. Furthermore, the dimensional proportions in the drawings are exaggerated for the convenience of explanation and may differ from the actual proportions.

[0030] (Overall composition) FIG. 1 is a diagram showing the overall configuration of an information processing system 1 according to this embodiment. The information processing system 1 includes an information processing device 10, which is a service management server, a terminal device 20, and a monitoring support system 30 used in each of multiple nursing care facilities 900. These devices communicate with each other via the Internet. The monitoring support system 30 includes devices present in the nursing care facility 900 (hereinafter simply referred to as "devices of the nursing care facility 900"). The information processing device 10 and the terminal device 20 are managed or used by a service provider. The service provider is an operating company that provides the monitoring support service of the monitoring support system 30, which will be described later. Furthermore, the operating body of the nursing care facility 900 (a medical corporation or a business that provides services covered by nursing care insurance) is the user of the monitoring support service.

[0031] In the following, an example will be shown in which one monitoring support system 30 is provided in each nursing care facility 900, but this is not limiting, and one monitoring support system 30 may be provided in multiple nursing care facilities 900. For example, in cases where an operating body operates multiple geographically different nursing care facilities 900, one monitoring support system 30 controls devices in the multiple nursing care facilities 900. Furthermore, the information processing system 1 may be composed of one information processing device 10 and one monitoring support system 30, and in this case, the information processing device 10 may be configured integrally with a monitoring device 32 of the monitoring support system 30 (see FIG. 5 etc. described below).

[0032] 2 is a block diagram showing a schematic configuration of the information processing device 10 according to the first embodiment. The information processing device 10 includes a control unit 110, a communication unit 120, and a storage unit .

[0033] (control unit 110) The control unit 110 includes a CPU, RAM, ROM, etc. The control unit 110 functions as an acquisition unit 111, an analysis unit 112, and an output unit 113, either alone or in cooperation with a communication unit 120.

[0034] The acquisition unit 111 acquires operation log data and response information log data from the watching support system 30. The operation log data and response information log data are stored in the storage unit 130. Specific examples of these log data will be described later.

[0035] The analysis unit 112 analyzes the operating status or usage status, including the utilization status, of the monitoring support system 30 at each nursing care facility 900 based on the operation log data and response information log data accumulated in the storage unit 130. The analysis unit 112 also analyzes changes over time in the usage status of the monitoring support system 30 at each nursing care facility 900 based on the operation log data and response information log data for a predetermined period accumulated in the storage unit 130. The predetermined period is a quarter, a month, or a week, and the status of the current period is compared with the status of the immediately preceding period, or the status of the current period is compared with the period at the time of introduction (start of operation). For example, if the predetermined period is one month, the analysis unit 112 calculates the average daily usage value for this month and the average daily usage value for the previous month (or for the month at the time of insertion). The analysis unit 112 then calculates the rate of change (rate of increase or decrease) for the current period relative to the past period. The one month period at the time of introduction (hereinafter also referred to as immediately after introduction) is, for example, one month period from the time when the first two weeks have passed (weeks 2 to 6).

[0036] The output unit 113 outputs the analysis result generated by the analysis unit 112. For example, it transmits it to the terminal device 20 of the service provider. By referring to the analysis result displayed on the terminal device 20, an employee 76 of the service provider determines whether the nursing care facility 900 (operating body) is utilizing the monitoring support service as intended, and considers the possibility of this operating body withdrawing from the monitoring support service.

[0037] (Communication unit 120) The communication unit 120 is an interface for transmitting and receiving data to and from each device via wired or wireless communication over the Internet.

[0038] (Storage unit 130) Various types of data such as service-related data are stored in the storage unit 130. The service-related data includes operation log data, response information log data, churn probability analysis results, service contract information, and the like.

[0039] 3A and 3B show examples of operation log data and response information log data. The contents of these will be described later. The service contract information is contract information regarding the services provided by the service provider to the operating body (user) of each nursing care facility 900. The contract information includes information regarding the number of devices installed in the nursing care facility 900 (the number of rooms in which the detection unit 31 described below is installed). Depending on the contract information, for example, the service provider will perform monthly or yearly billing processing for the user and provide the corresponding nursing care-related monitoring support service. The churn probability analysis result is the analysis result of the churn probability analysis processing (FIG. 16, etc.) described later, and is the analysis result of the possibility of the user churn from the monitoring support service or the possibility of the user using the device of the nursing care facility 900 churn.

[0040] (Terminal device 20) 4 is a block diagram showing a schematic configuration of the terminal device 20 (which functions as a display device). The terminal device 20 is a PC (personal computer) and includes a control unit 210, a communication unit 220, a display unit 230, and an input unit 240, which are interconnected by a bus.

[0041] The control unit 210 includes a CPU, RAM, ROM, etc. The control unit 210 may also include an HDD (hard disk drive). The communication unit 220 is an interface for various local connections, such as a network interface for wired communication according to standards such as Ethernet (registered trademark), or an interface for wireless communication according to standards such as Bluetooth (registered trademark) or IEEE802.11, and communicates with devices connected via a LAN.

[0042] The display unit 230 is, for example, a liquid crystal display, and displays various information. The display unit 230 displays, for example, the analysis results provided from the output unit 113 of the information processing device 10.

[0043] The input unit 240 includes a keyboard, a numeric keypad, a mouse, etc., and is used to input various types of information.

[0044] (Monitoring Support System 30) The configuration of the watching support system 30, the operation log data of the watching support system 30, and the response information log data will be described below with reference to FIGS.

[0045] 5 is a diagram showing the overall configuration of the watching support system 30. The watching support system 30 used in the nursing care facility 900 includes a detection unit 31, a terminal device 33, a staff terminal 34, and a wireless AP (access point) 35.

[0046] The detection unit 31 is placed in each room (hereinafter also referred to as living room) of the resident 70, and detects the behavior and state of the resident 70 in the room. The staff terminal 34 is a portable terminal such as a smartphone, and is used by the staff 72, such as caregivers or nurses, who provide care and nursing for the resident 70.

[0047] The terminal device 33 is a PC and is used by an administrator 74. The administrator 74 is, for example, a manager or representative of the nursing care facility 900, or a leader of the staff 72, and performs notification settings, which will be described later. Devices such as the detection unit 31, monitoring device 32, and terminal device 33 are communicatively connected to one another via a LAN. The staff terminal 34 is connected to other devices via a wireless AP 35. Furthermore, each device in the nursing care facility 900 is connected to the Internet via an RT (router).

[0048] (Detection unit 31) FIG. 6 is a diagram showing an example of a detection unit 31 installed around a bed 60 in a room of a resident 70. A detection unit 31 (also referred to as a sensor box) is placed in each room so as to monitor the room of the resident 70, which is the observation area. The detection unit 31 includes various sensors as described below, and generates sensing information from the output of the sensors. The sensing information includes not only signals from each sensor, but also, for example, a determination result of a person's movement determined by image processing from a video signal (an event as described below), and a care call signal from a care call unit 314 (described below).

[0049] Fig. 7 is a block diagram showing a schematic configuration of the detection unit 31. As shown in Fig. 7, the detection unit 31 includes a control unit 311, a communication unit 312, a camera 313, a care call unit 314, a voice input / output unit 315, and a pause switch 316, which are interconnected by a bus or short-range wireless communication.

[0050] The control unit 311 has the same configuration as the control unit 210, including a CPU, RAM, ROM, etc. The control unit 311 may also have an HDD. The control unit 311 controls each unit of the detection unit 31 and performs calculation processing according to a program. The control unit 311 functions as an event determination unit 31a.

[0051] The communication unit 312 is an interface circuit (for example, a LAN card, a wireless communication circuit, etc.) for communicating with other devices such as the monitoring device 32 via a LAN.

[0052] The camera 313 is placed, for example, on the ceiling or upper part of a wall of the room, captures an image of the bed 60 of the resident 70 directly below as an observation area, and outputs the captured image (image data). This captured image includes still images and videos. The camera 313 is a near-infrared camera, but a visible light camera may be used instead, or a combination of these may be used. The camera 313 outputs an image signal as sensing information acquired by the imaging sensor.

[0053] The care call unit 314 includes a push-button switch, and detects a care call (also called a nurse call) when the resident 70 operates this switch. Instead of a push-button switch, a care call may be detected by an audio microphone. When the switch of the care call unit 314 is pressed, that is, when a care call is detected, the control unit 311 transmits a notification that a care call has been made to the monitoring device 32, etc., via the communication unit 312 and the LAN.

[0054] The audio input / output unit 315 is, for example, a speaker and a microphone, and enables audio communication by transmitting and receiving audio signals to and from the terminal device 33 and the staff terminal 34 via the communication unit 312. The audio input / output unit 315 may be connected to the detection unit 31 via the communication unit 312 as an external device of the detection unit 31.

[0055] Pause switch 316 is placed near the entrance to the room. This pause switch 316 includes a push button switch, and when staff member 72 or the like operates this switch when entering the room, a pause signal is sent to control unit 311. In response to this pause signal, the determination process by event determination unit 31a of detection unit 31, which will be described below, is temporarily stopped.

[0056] In the event determination process described below, the event determination unit 31a of the control unit 311 determines an event based on the movement of the resident 70 based on the captured image. In this case, if a third party other than the resident 70, such as a staff member 72, enters the room, the control unit 311 may not be able to distinguish between the resident 70 and the third party and may interpret the third party's actions as those of the resident 70, resulting in an unintended event determination. In other words, there is a risk of a false alarm. To prevent this, a pause switch 316 is used. Note that the pause switch 316 may be a push button type or may also be an NFC (near field communication) sensor such as FeliCa (registered trademark). For example, the pause switch is turned on / off by holding an IC card held by the staff member 72 over a reading terminal installed at the entrance / exit of the room. Operating the pause switch temporarily suspends the event determination process, but this suspension can be released by operating the pause switch again. Note that the pause switch may be automatically released after a predetermined time has elapsed, according to a setting previously made by the staff member 72 or the like.

[0057] The detection unit 31 may further include other sensors for detecting nursing care data, such as a body movement sensor, a bed sensor, a mat sensor, a thermal sensor, or an infrared sensor. This body movement sensor may be a Doppler shift sensor that transmits and receives microwaves to and from the bed 60 and detects the Doppler shift of the microwaves caused by the body movement (e.g., breathing) of the resident 70. This body movement sensor detects chest movement (up and down movement of the chest) caused by the breathing of the resident 70. If a disruption in the period of the chest body movement or an amplitude of the chest body movement below a predetermined threshold is detected, it is determined to be an abnormal micro-body movement (e.g., due to cardiac arrest). This determination of abnormal micro-body movement may be made by the detection unit 31, or may be made by the control unit 110 of the information processing device 10 by sending only a signal to the information processing device 10. This bed sensor may be any sensor that can be attached to a bed, and may, for example, detect weight and be placed on the bed 60 or on the floor at the exit of the bed 60 as the observation area, and may detect whether or not a person is standing on the sensor, or may detect the state of sleep. This infrared sensor is also called a human presence sensor, and detects whether or not a person is present in the room. For example, it may be placed throughout the room or on the bed 60 as the observation area.

[0058] (Event detection by image processing) In this embodiment, as described above, an event is a change in the state of the resident 70 detected by the detection unit 31, such as getting up, getting out of bed, falling, or abnormal slight body movement, which requires care for the resident 70 and for which an alert (notification) should be issued to the staff 72. This event also includes a care call made by the care call unit 314. The detection unit 31 transmits the determination result of the event that has occurred to the monitoring device 32.

[0059] The event determination unit 31a of the control unit 311 detects the behavior of the resident 70 from the images captured by the camera 313. The behaviors to be detected include "getting up" which is getting up from the bed 60, "getting out of bed" which is getting away from the bed 60 after getting up, "falling" which is falling off the bed 60, and "falling down" which is falling onto the floor or the like.

[0060] The event determination unit 31a detects image silhouettes (hereinafter referred to as "human silhouettes") from multiple captured images (moving images). Human silhouettes can be detected, for example, by extracting a range of pixels with relatively large differences using time subtraction, which subtracts images captured at different times. Human silhouettes may also be detected using a background subtraction method, which subtracts a background image from a captured image. When a human silhouette is detected, it is determined that a resident 70 is present in the room.

[0061] Furthermore, the area within the rectangle circumscribing the outline of the image of the resident 70 can be detected as a human rectangle circumscribing the outline of the image of the resident 70. Furthermore, together with or instead of a human silhouette obtained from the overall image of the person, a silhouette of the image of the head of the resident 70 (hereinafter referred to as a "head silhouette") may be detected. The area within the rectangle circumscribing the outline of the image of the head of the resident 70 can be detected as a head rectangle circumscribing the outline of the image of the head of the resident 70 in the captured image.

[0062] The detection of getting up, getting out of bed, falling, and tipping over is performed by detecting the posture of the resident 70 (e.g., standing, sitting, lying down, etc.) from the detected human silhouette and their relative position to objects in the room, such as the bed 60, to determine whether the event is getting up, "getting out of bed," or "falling over." For example, "getting up" can be recognized when the width of the human silhouette crossing any side of a predetermined rectangle, viewed from above, with the four corners of the bed 60 as vertices, increases to 20 cm or more. Furthermore, "getting out of bed" can be recognized when the ratio of the silhouette's area outside the rectangle to the area inside the rectangle increases to 80% or more. Furthermore, when determining the sitting position, it may also be recognized whether the sitting position is near the edge of the bed 60 (sitting on the edge of the bed 60). When the event determination unit 31a of the detection unit 31 detects any of these types of events, such as getting up, getting out of bed, or falling over, it transmits event information indicating the occurrence of the event to the monitoring device 32, etc.

[0063] Such event determination processing by the event determination unit 31a may be performed by a program processed by the CPU of the control unit 311, or by an embedded processing circuit. As another example, all or most of the detection processing may be performed on the monitoring device 32 (control unit 321) side as described above. In this case, the control unit 311 of the detection unit 31 only transmits the captured image (data related to nursing care) to the monitoring device 32, and the monitoring device 32, upon receiving this, detects an event based on the captured image.

[0064] Fig. 8 is a block diagram showing a schematic configuration of the watching device 32. The watching device 32 includes a control unit 321, a communication unit 322, and a storage unit 323. The watching device 32 is a server (edge ​​server) and provides a watching support service by working in cooperation with the information processing device 10. Note that the device configuration shown in Fig. 8 and other figures is merely an example, and some of the functions of the watching device 32 may be performed by the information processing device 10.

[0065] (Control unit 321) The control unit 321 includes a CPU, RAM, ROM, etc. The control unit 321 functions as a notification unit 32a and a notification setting unit 32b, either alone or in cooperation with the communication unit 322. When the notification unit 32a receives event information from the detection unit 31, it notifies the staff 72 in charge of caring for the resident 70 who caused the event (see step S150 in FIG. 11, which will be described later). The notification setting unit 32b sets notification ON / OFF for each room (each observation area) according to settings made by a user (e.g., an administrator 74). If notification is set to OFF, the notification unit 32a does not notify the staff 72. The notification setting procedure will be described later (see FIGS. 14 and 15, which will be described later).

[0066] The communication unit 322 is an interface circuit (for example, a LAN card, a wireless communication circuit, etc.) for communicating with other devices such as the detection unit 31 via a LAN.

[0067] (Storage unit 323) The memory unit 323 stores a resident list, a staff list, an event list, and shift information.

[0068] The "resident list" and "staff list" include personal IDs (names and numbers), room numbers, staff names, assigned units, etc., for residents 70 and staff 72. A unit (also called a responsible group) is a resident unit to which multiple residents who share a common area, such as the same floor, belong within the facility, and a group to which multiple staff members who are responsible for these resident units belong. The "event list" (see Figure 12, described below) includes event information and response status. "Shift information" is the working hours or working period of each staff member 72, and is recorded in accordance with the login / logoff times, as described below. Care records may also be stored. The "care records" include care record sheets, nursing record sheets (medical charts, life records), and diagnostic charts, which record the care provided to these residents 70 by staff members 72 (caregivers, nurses, etc.).

[0069] (Terminal device 33) 9 is a block diagram showing a schematic configuration of the terminal device 33. The terminal device 33 is a PC and includes a control unit 331, a communication unit 332, a display unit 333, and an input unit 334. These components are similar to those of the terminal device 20, and therefore a description thereof will be omitted. A facility manager 74 performs notification settings, which will be described later, via the terminal device 33.

[0070] (Staff terminal 34) FIG. 10 is a block diagram showing the schematic configuration of the staff terminal 34. The staff terminal 34 includes a control unit 341, a wireless communication unit 342, a display unit 343, an input unit 344, and an audio input / output unit 345, all of which are interconnected via a bus. The control unit 341 includes a CPU, RAM, ROM, and other components similar to the control unit 311 of the detection unit 31. The wireless communication unit 342 enables wireless communication using standards such as Wi-Fi and Bluetooth (registered trademark), and wirelessly communicates with each device directly or via the wireless AP 35. The display unit 343 and the input unit 344 are touch panels, with a touch sensor superimposed on the display surface of the display unit 343, which is typically a liquid crystal display. The display unit 343 and the input unit 344 display various operation screens (see FIG. 13, described below) that display a list of multiple events included in the event list to the staff member 72, and accept various operations through these operation screens. The accepted operations are sent to the monitoring device 32 as operation information (response information). The voice input / output unit 345 is, for example, a speaker and a microphone, and enables the staff member 72 to make voice calls to other staff terminals 34 via the wireless communication unit 342. The staff terminal 34 can be configured as, for example, a portable communication terminal device such as a tablet computer, a smartphone, or a mobile phone.

[0071] At the start and end of work, the staff member 72 performs login authentication processing and logout processing, respectively, through the staff member terminal 34 assigned to them. These periods and times are stored in the memory unit 323 as the shift information described above. The staff member 72 inputs a staff member ID and password through the touch panel (display unit 343, input unit 344) of the staff member terminal 34 and transmits them to the monitoring device 32. The monitoring device 32 compares the authentication information stored in the memory unit 323 and transmits an authentication result according to the authority of the staff member 72 to the staff member terminal 34, thereby completing the login authentication.

[0072] Each staff member 72 belongs to the unit to which they belong. For example, 24-hour care is provided to multiple residents 70 (care recipients) belonging to a resident unit in the East Wing on the second floor of a nursing facility by multiple staff members belonging to the same unit. In this case, if an event related to a resident 70 is detected in a resident unit in the East Wing on the second floor, the event information is broadcast to the staff terminals 34 of all staff members (who are logged in) belonging to the unit in charge of that resident unit, as will be described later. The resident units and the names of the members of the units to which they belong are registered in the resident list and the staff list, respectively.

[0073] (Event handling processing) FIG. 11 is a flowchart showing the process of responding to an event related to the watching support service executed by the watching support system 30.

[0074] (Step S100) Here, a login authentication process is performed. The staff member 72 inputs a staff member ID and password via the touch panel (display unit 343, input unit 344) of the staff member terminal 34 and transmits them to the watching device 32. The watching device 32 compares the authentication information stored in the memory unit 323 and transmits the authentication result according to the authority of the care staff member to the staff member terminal 34.

[0075] (Step S110) The detection unit 31 monitors the observation area with a camera 313 .

[0076] (Steps S120, S130) The detection unit 31 detects an event from the movement of the resident 70 in the observation area. The detection unit 31 also detects an event through the operation of the care call unit 314. Then, the communication unit 312 (acquisition unit) of the watching device 32 acquires event information of the event detected by the detection unit 31.

[0077] (Step S140) The monitoring device 32 updates the event list based on the event information acquired in step S130. FIG. 12 shows an example of an event list stored in the storage unit 323. Events included in the event list include, as described above, getting up, getting out of bed, falling, tipping over, and care calls (also referred to as care call events). The most recently detected event is added to the end of the event list. Each event includes "event information" and "response status" data. The event information includes an automatically assigned primary key event ID, a room number, a target person, an event type, the date and time of occurrence, and image data. The image data (photographed images i010 to i013) included in the event information is still image data captured by the detection unit 31 when the event occurred, and transmitted in step S130. The "response status" includes the status, the staff member in charge, the date and time of response, whether live images (hereinafter also referred to simply as videos) were viewed (the captured images were used), and whether the room was visited. The staff member in charge is the staff member who took the initiative in the processing of step S180, described below, to assign themselves to the event. Here, "visiting a room" refers to the act of visiting the room of the resident 70 who caused the event to occur.

[0078] (Step S150) Here, the notification unit 32a of the monitoring device 32 notifies one or more staff members 72 of the event information newly acquired in step S120. Note that, with reference to the notification settings in the memory unit 323, if the room in which the event occurred has notifications OFF (all OFF) or if notifications for a specific event type (see FIG. 15(b) described below) are set to OFF, notification of the corresponding event is not made. If notifications are set to ON, a broadcast notification is made to simultaneously notify the staff terminals 34 of all logged-in staff members 72 belonging to the unit to which the resident 70 belongs, based on the event information. If notifications are OFF and no notification is made, the number of notifications is not counted. If notifications are ON and a notification is made, the number of notifications is counted as one. Note that FIG. 11 only shows the processing of one staff member terminal 34 used by one staff member 72 out of multiple staff members 34.

[0079] (Steps S160, S170) Each staff member 72 checks the event that has occurred through the staff terminal 34. Fig. 13 shows examples of operation screens d1 and d2 for responding to the event notification (broadcast notification) displayed on the display unit 343 of the staff terminal 34. The operation screen d1 shown in the figure is a screen that displays a list of events. The displayed events can be scrolled up and down by flicking.

[0080] Areas a10 to a12 on the operation screen d1 correspond to event IDs 010 to 013 in the event list of FIG. 12. Area a10 displays an icon indicating that event ID 010 is a care call and that the event has been addressed. Area a11 displays an icon indicating that event ID 011 is a wake-up event and that the event has been addressed. Area a12 displays icons indicating that event IDs 012 and 013 are a wake-up event and a bed-getting event and that the event has not yet been addressed. These two events are related to the same resident 70 (Mr. C), so they are displayed together in the same area. Furthermore, because the event has not yet been addressed, area a13 displays a thumbnail image (image i013) related to the most recent event (ID 013). By operating area a13, text messages can be exchanged with other staff members 72 who are logged in. By clicking area a12 on the operation screen d1, the staff member 72 transitions to operation screen d2, which is used to confirm the next event.

[0081] (Step S170) The staff member 72 checks the details of the event on the operation screen d2 displayed on the staff terminal 34. On the operation screen d2, the name of the resident (Mr. C) who caused (determined) the event and the name of the staff member (Staff Member D) using the staff terminal 34 are displayed in area a20. The time elapsed since the event occurred is displayed in area a21. The elapsed time ("1 minute elapsed") shown in area a21 is the time elapsed since the most recent event was determined (rounded down to the nearest minute). Area a21 also displays a thumbnail image (still image) of the image captured at the time of the event (leaving bed). If the thumbnail image is insufficient and the staff member 72 wants to further confirm the status of the resident 70 who caused the bed leaving or getting up event, the staff member 72 operates the "talk" or "view" buttons in areas a23 and a24 below. Operating the "talk" button allows the staff member 72 to talk to the resident 70 (Mr. C) via the audio input / output unit 315. By operating the "Watch" button, a display request is sent to the monitoring device 32, and in response, live images (real-time captured images (video)) captured by the camera 313 can be viewed by streaming playback on the operation screen d2. When the "Watch" button is operated, as shown in the video viewing column of the event list in FIG. 12, video viewing "Yes" is recorded for this event (hereinafter also referred to as use of the "Live video function"). At this time, the video viewing time and the number of calls / call time via the "Talk" button may also be recorded in association with the event.

[0082] A staff member 72 (e.g., staff member D) operates the "Respond" button (hereinafter also referred to as "request for responsibility") in area a22 to indicate that he or she wishes to take the initiative in taking charge of the displayed events (ID012, 013) for getting up and getting out of bed. If he or she does not wish to take charge, he or she can return to the previous operation screen d1 showing the event list by operating the back button (triangle icon) in the bottom column. Note that while a staff member 72 is displaying the operation screen d2 shown in FIG. 13 and checking the status of the event, the monitoring device 32 controls the prohibition process for other staff members 72. For example, even if a staff member 72 is checking the status of an event for resident 70 (Mr. C) and another staff member 72 selects the same event (ID013), the words "Checking status" are displayed on the operation screen d1 for the other staff member 72, and the "Respond" button is not displayed or cannot be selected even when the operation screen d2 is transitioned to.

[0083] (Steps S180, S190) If staff member D wishes to take the initiative and take charge of an event, he or she operates the "take charge" button (presses the take charge button). In response to staff member D, who is the terminal user, operating the "take charge" button, the staff terminal 34 sends a request to take charge of the selected event to the monitoring device 32. In response, the monitoring device 32 replies with an approval notice. Upon approval, staff member D, who sent the request, takes charge of the event, checks the status of the event via the staff terminal 34, and, if necessary, rushes to the event to take charge (visit the room), etc.

[0084] (Step S200) The monitoring device 32 updates the event list in response to the processing of step S180. Specifically, the status of the event list (IDs 012 to 013: getting out of bed, etc.) is changed to "addressed." In this embodiment, the status is changed to "addressed" in response to the staff member 72 operating the "address" button for the event selected and displayed on the operation screen d2. That is, operating the "address" button inputs a response status of "addressed," which is information indicating confirmation, and the staff member 72 who operated the button is deemed to have responded, and the response date and time are recorded. More specifically, in the event list of FIG. 12, the time when the processing of step S180 or S190 was performed is recorded in the response date and time column. In this case, the staff member 72 may end the response without rushing to the event by checking the thumbnail image on the operation screen d2 or talking to the resident 70. However, the present invention is not limited to this, and the response date and time may be recorded in the response date and time field as the time when the staff member 72 rushes to the room of the resident 70 who has experienced an event such as a fall, a tumble, or a nurse call and determines that the staff member has visited the room (step S220 described below). As another example, the response result may be entered on a separate operation screen to mark the event as completed, and the time of entry may be recorded in the response date and time field. Furthermore, when the event list is updated, the monitoring device 32 broadcasts the updated event list to all staff members who are currently logged in.

[0085] (Step S210) Here, an example will be described in which an emergency response is performed. The staff member 72 in charge of the event visits the room of the resident 70 who caused the event to respond to the emergency response.

[0086] (Step S220) When the pause switch 316 is operated, or when the room detection unit 31 (event determination unit 31a) detects a third party other than the resident 70 based on the image captured by the camera 313, i.e., when it detects the presence of multiple people, it determines that a staff member 72 has visited (entered) the room. If a visit to the room where the event occurred occurs within a predetermined time (e.g., within 10 to 15 minutes) after the event response request (S180) was made, it is deemed that the staff member 72 who made the response request (staff member D) has visited the room. When an NFC-based reading terminal placed at the entrance / exit of the room is used as the pause switch 316, the staff member 72 may hold an IC card with ID data recorded on it over the reading terminal to determine that a visit has occurred and which staff member 72 has visited the room.

[0087] (Step S230) The detection unit 31 transmits information about the staff member's visit to the monitoring device 32. The visit information includes the resident's 70 ID number.

[0088] (Step S240) The monitoring device 32 records the room visit information received in step S230 in association with the event for which the status of the resident 70 corresponding to the ID number is "completed" and which has the latest completed date and time. For example, the monitoring device 32 updates the "visited" column of the completed status of the corresponding event ID in the event list of FIG. 12 to "yes."

[0089] (Step S250) When the staff member 72 finishes working, the staff member 72 logs out by operating a "work end" button (not shown) on the staff member terminal 34. The logout time is recorded in the storage unit 323 as shift information.

[0090] (Notification settings) Next, notification settings will be described with reference to FIGS. 14 and 15. FIG. 14 shows an example of an operation screen d3 for setting notification, and FIG. 15 shows notification settings and types stored in the storage unit 323. The operation screen d3 in FIG. 14 is displayed on the display unit 333 of the terminal device 33. The administrator 74 can set notifications ON / OFF by clicking a bell icon i01 provided corresponding to each room on the operation screen d3. The notification settings are stored in the storage unit 323. In the operation screen d3 in FIG. 14, rooms 101, 104, 105, etc. are set to notification OFF and displayed as grayed out. The other rooms (rooms 102, 103, etc.) are set to notification ON. The notification settings are stored in the storage unit 323 in the format shown in FIG. 15(a). Note that the example in FIG. 14 shows an example in which two settings, ON / OFF, are performed, but multiple settings may also be performed. FIG. 15(b) shows an example in which multiple settings are performed. Notifications for each event of getting up, getting out of bed, and falling can be individually set to ON / OFF. For example, in notification setting 7, notifications for getting up and getting out of bed are OFF, and notifications for falling are ON. In this case, the responsible staff member 72 is notified of the event of falling, but notifications for other events such as getting up and getting out of bed are not sent. For example, for a resident 70 with a low level of care needs and who can walk independently and steadily, notification setting 8 (all OFF) is set; conversely, for a resident with a high level of care needs and who has difficulty walking independently, notification setting 1 (all ON) or an intermediate notification setting such as 5 (notifications for getting up OFF) is set.

[0091] Furthermore, the operation screen d3 displays the state (state, respiratory rate, heart rate) of the resident 70 obtained from the sensing information of the detection unit 31 so that the staff 72 or manager 74 can grasp the situation in each room at once. Based on the sensing information of the camera 313, it is shown that room 102 is in an awake state (wake-up event), room 202 is in a sitting state, room 301 is in an out-of-bed state (bed-out event), and room 304 is in a sleeping state. Furthermore, when the resident is asleep or awake and lying down in bed 60 (room 102, etc.), the heart rate and respiratory rate are displayed based on the sensing information of the body movement sensor.

[0092] (Operation log data and response information log data) In this way, the monitoring device 32 acquires sensing information related to nursing care from each device (device present in the nursing care facility 900) in the nursing care facility 900, generates operation log data and response information log data from this information, and stores them in the storage unit 323. Please refer to Figures 3A and 3B again.

[0093] (Operation log data) The operation log data is data that indicates the operation status of each device in the monitoring support system 30. As shown in Fig. 3A, the operation log data includes notification settings for each room (x1), event notifications (x2), and login / logout (x3).

[0094] x1: Notification setting is an ON / OFF setting of event notification for each room, set by the notification setting unit 32b as shown in FIGS.

[0095] x2: Event notification corresponds to step S130 (or step S150) in Fig. 11 described above. This is a notification to the staff terminal 34 regarding an event requiring care for the resident 70, which was detected based on sensing information from the camera 313, the event determination unit 31a, the care call unit 314, and the notification unit 32a. This notification is associated with event information (event ID, event type, date and time of occurrence, etc.) and recorded in the event list.

[0096] x3: Login / logout corresponds to steps S100 and S250 in Fig. 11 described above, and is performed by the staff terminal 34. This signal corresponds to the start and end times of work of the staff member 72 who uses this staff terminal 34, and is stored in the memory unit 323 as shift information.

[0097] (Response information log data) The response information log data is information about responses to events requiring care for residents by staff 72 using the monitoring support system 30. As shown in FIG. 3B, the response information log data includes response button press information (y1), pause signal (y2), and room visit information (y3).

[0098] y1: The response button press information corresponds to step S180 in Fig. 11 described above, and is performed by the staff terminal 34. The response button press information is a "response request" from the staff member 72, and is recorded in the event list in association with the event information.

[0099] y2: The pause signal is output in response to operation of the pause switch 316, and is recorded in the event list in association with the event information.

[0100] y3: Room visit information corresponds to step S220 in Fig. 11. The fact that the staff member 72 who issued the event response request has visited the room is recorded in the event list as room visit information through an image captured by the camera 313 or operation of the pause switch 316.

[0101] (Analysis of the likelihood of withdrawal) Next, the process of analyzing the churn possibility based on the above-mentioned operation log data and response information log data will be described with reference to Fig. 16. Fig. 16 is a flowchart showing the churn possibility analysis process.

[0102] (Steps S41 and S42) The acquisition unit 111 acquires operation log data and response information log data (see FIGS. 3A and 3B) from each monitoring support system 30 of each care facility 900. The control unit 110 stores the acquired operation log data and response information log data in the storage unit 130.

[0103] (Step S43) The control unit 110 determines whether it is time to perform analysis. For example, it determines that analysis should be performed every predetermined day, such as once a month, or when a request is received from the terminal device 20 (employee 76) (YES), and proceeds to step S44; otherwise (NO), it repeats the processing from step S41 onwards.

[0104] (Step S44) The analysis unit 112 analyzes the possibility of leaving one or more nursing care facilities 900 for which the analysis timing has arrived. For the process of analyzing the possibility of leaving, the subroutine process shown in FIG.

[0105] (Step S501) 17, the analysis unit 112 refers to the operation log data and the response information log data, and calculates a plurality of indicators of withdrawal signs by evaluating the operation or utilization status of the watching support system 30 or the watching support service. Fig. 18 is a table showing examples of the indicators of the likelihood of withdrawal calculated in step S501.

[0106] (Indicator of likelihood of churn (usage status)) The analysis unit 112 of the information processing device 10 calculates (quantifies) at least one of the withdrawal possibility indices i1 to i12 shown in Fig. 18 based on the operation log data and the corresponding information log data accumulated in the storage unit 130 by the processing of step S42. The indices i1 to i12 are indices related to the usage status of the watching support system 30, and among these, the indices i1 to i3 relate to the operation status, and the indices i4 to i12 relate to the utilization status.

[0107] Indicator i1 (rate of increase / decrease in the number of notifications): The analysis unit 112 calculates the rate of increase / decrease in the number of notifications using operation log data (x2: event notifications). The number of notifications is calculated by calculating the number of notifications this month (per day, per room) and the same number of notifications last month over a predetermined period (one month). Then, the ratio of the number of notifications this month to the number of notifications last month is calculated to analyze changes over time. Furthermore, if there is an increase / decrease that exceeds a certain range, for example, if the range is 50 to 200%, and it is less than 50% or more than 200%, the analysis unit 112 will make an alert judgment (hereinafter also referred to as an NG judgment).

[0108] Indicator i2 (proportion of all notifications ON): The analysis unit 112 calculates the proportion of all notifications ON (notification ON rate) based on the operation log data (x1: notification settings) and the service contract information (number of rooms in which the detection unit 31 (sensor) is installed). Furthermore, if the recent (current) notification ON rate exceeds a certain value or is less than a certain value, the analysis unit 112 may determine that the monitoring support service is not being used appropriately and may therefore be judged as NG.

[0109] Indicator i3 (notification OFF rate): This notification OFF rate is used when notification OFF can be set in multiple stages, as in the example shown in FIG. 15(b). The analysis unit 112 calculates the ratio of the number of notification OFF settings to the setting value (number of rooms × number of notification types) obtained by multiplying the number of rooms in which the detection unit 31 is set by the number of notification OFF setting types (3). For example, in the example shown in FIG. 15(b), for notification setting 5, the number of notification OFF settings in one room is 1 (only wake-up is OFF), for notification setting 7 it is 2, and for notification setting 8 it is 3. Similarly, if the recent (current) notification OFF rate exceeds or is below a certain value, the analysis unit 112 determines that the monitoring support service is not being used appropriately and judges it to be NG.

[0110] Indicator i4 (response time increase / decrease rate): The analysis unit 112 calculates the response time using operation log data (x2 (event notification), x4 (response button press information)). This response time corresponds to the elapsed time from step S150 (notification) to step S180 (handling request) in FIG. 11. The analysis unit 112 calculates the average response time (per notification) for this month and the average response time for the previous month for a predetermined period (one month). Then, the analysis unit 112 analyzes changes over time by calculating the ratio of this month's average response time to the previous month's average response time. If there is an increase or decrease that exceeds a certain range, for example, if the range is 50 to 200%, the analysis unit 112 determines that the increase or decrease is NG if it is less than 50% or more than 200%. For example, if the increase or decrease is extremely short, it is suspected that the response button is simply being pressed without checking the content of the event, and if the increase or decrease is extremely long, it is suspected that the response button has stopped being used.

[0111] Index i5 (notification-induced visit rate): Calculated by dividing the number of visits (a) made by staff 72 to the room of a resident involved in an event notification by the number of event notifications (b) (step S150 in FIG. 11). Visit rate i5 = number of visits (a) / number of notifications (b). The analysis unit 112 may also determine that the visit rate i5 is not acceptable if it is extremely high or outside a predetermined range. This is because the live video function or call function (corresponding to areas a23 and a24 on the operation screen d2 in FIG. 13) may not be fully utilized. Furthermore, a monthly change in the visit rate i5, i.e., a comparison of the visit rate (per day) for the current month and the previous month, shows a continuous decrease of a certain value or more, and may also result in a negative decision.

[0112] Index i6 (event notification response rate): Calculated by dividing the number c of times the staff member 72 pressed the response button in response to an event notification (corresponding to the handling request in steps S150 and S180 of FIG. 11) by the number b of event notifications. Response rate i6 = number of presses c / number of notifications b. The analysis unit 112 may also reject responses outside a predetermined range, such as when the response rate i6 is extremely high. This is because the response button may be pressed perfunctorily (perfunctorily) without checking the event content. The analysis unit 112 also calculates the monthly trend of the response rate i6, i.e., by comparing the visit rate (per day) for the current month with the previous month. A consecutive decrease of a certain value or more may result in a rejection. The events used to calculate the response rate i6 may be calculated based on all events, including care call events, or on events other than care calls.

[0113] Index i7 (care call notification response rate): The response rate is calculated based on care call events. Response rate i7 = number of presses c / number of care call notifications d. Except that the event type is limited to specific care calls, the calculation method and NG judgment are the same as for response rate i6. Note that care call unit 314 may be operated continuously (by repeated taps), and if the time interval between previous and next care calls is within a predetermined time (for example, within one second), it is preferable to remove and count the second and subsequent care calls. This removal process can be performed on the detection unit 31 side (for example, the process of step S120).

[0114] Indicator i8 (Live video function usage rate): This is calculated by dividing the number of times e the Live video function was viewed by the number of times b the event was notified when checking the event details in step S170 of Fig. 11. Usage rate i8 = number of times e used / number of times b notified.

[0115] Indicator i9 (average video viewing time): The average video viewing time per session when using the Live video function.

[0116] Indicator i10 (notification effective utilization rate): This is calculated using the number of event notifications b, response rate i6, and visit rate i5. Utilization rate i10 = number of notifications b × response rate i6 × visit rate i5. As with visit rate i5 and response rate i6, the utilization rate i8, confirmation time i9, and utilization rate i10 may also be judged as NG if their values ​​are outside a specified range, or if they are consistently outside the specified range over a monthly period.

[0117] Indicator i11 (decrease rate in number of visits): This uses the number of visits a itself. The analysis unit 112 calculates the monthly decrease rate by dividing the number of visits a this month by the number of visits a in January immediately after implementation. The decrease rate in number of visits is an indicator of work efficiency. For example, by using a live video function or a call function to check the situation and then omitting a visit, that time can be allocated to other work. If the decrease rate is lower than a specified range (if the number of visits remains high), it may be judged as NG. In this case, it can be assumed that the monitoring service is not being used effectively.

[0118] Indicator i12 (rate of increase or decrease in working hours): The analysis unit 112 calculates the total working hours of the staff 72 based on the shift information. This working hour may be normalized by the number of staff 72 working at the nursing care facility 900 or the number of residents 70. The analysis unit 112 calculates the monthly decrease rate by dividing the working hours f of this month by the working hours f of January immediately after the introduction of the service. The rate of increase or decrease in working hours is an indicator of work efficiency. For example, if the working hours have increased or if there has been no change since the introduction of the service, it may be determined that the service is not working properly. In this case, it can be assumed that the monitoring service is not being used effectively. Note that, with regard to the indicators i1, i5 to i12, the analysis unit 112 analyzed the changes over time on a monthly basis. Alternatively, the analysis unit 112 may analyze the changes over time on a weekly, quarterly, or yearly basis.

[0119] (Step S502) Referring again to Fig. 17, the analysis unit 112 calculates the indices for each group as shown in Fig. 19 and Fig. 20. Fig. 19 is a table showing the indices classified into factors 1 and 2 used to determine the stage, and Fig. 20 is a table showing the indices classified by phase.

[0120] Referring to Figure 19, factor 1 is a set of elements related to the likelihood of withdrawal, classified from the perspective of whether users who have introduced the monitoring support system (or monitoring support service) are unable to achieve their purpose for introducing the system, and the corresponding indicators related to the likelihood of withdrawal. Factor 2 is the same elements and indicators classified from the perspective of users' dissatisfaction with the monitoring support system. Each indicator corresponds to indicators i1 to i12 shown in Figure 18 above.

[0121] Factor 1, "The purpose of implementation cannot be achieved," includes factors such as "working hours do not decrease," "staff workload does not decrease," "turnover rate," "total working hours at the facility," and "staffing allocation." Factor 2, "dissatisfaction with the system," includes factors such as "not understanding how to use it" and "defects." The analysis unit 112 judges items e1 to e12 corresponding to each factor. The monthly working hours per person for item e1 corresponds to indicator i12. If the monthly working hours are not within the predetermined range (determination threshold) of 150 hours or less, it is judged to be outside the predetermined range (corresponding to the alert or NG judgment described above). Furthermore, for item e2, if the ratio of monthly working hours per person for the current month to the month immediately after implementation is not within the predetermined range of 95% or less, i.e., if it exceeds 95%, it is judged to be NG. A similar judgment is made for items e3 to e12. For items e5 (turnover rate) and e7 (staffing increase / decrease rate), the acquisition unit 111 periodically acquires the staff list and resident list stored in the storage unit 323 and stores them in the storage unit 130. For item e5, the analysis unit 112 determines the turnover rate for the target nursing care facility 900 by comparing the list from a predetermined period ago (e.g., one year ago) with the current list. If the turnover rate is not within the predetermined range (60% or more), the analysis unit 112 makes an NG determination. For item e7, the analysis unit 112 calculates the increase / decrease rate of the current number of residents 70 and staff 72 for the target nursing care facility 900. This increase / decrease rate is calculated from the number of staff 72 in the current month relative to the number of staff 72 in the introduction month. This number of staff 72 corresponds to the allocation ratio and may be calculated from the total working hours of the staff 72 and the number of staff (persons / 8 hours) converted from the total number of residents and working hours.

[0122] Referring to Fig. 20, the phases are classified into "Trial Version," "Official Version Launch," and "Stable Operation" according to the time period, and the indicators corresponding to each phase correspond to indicators i1 to i12 shown in Fig. 18. As with item numbers e1 to e12 in Fig. 19, item numbers f1 to f15 in Fig. 20 are also compared with a predetermined range (determination threshold), and a determination is made as to whether or not each item is NG. Note that the variations in item numbers f14 and f15 relate to the variations in the number of responses to notifications by each staff member 72 in the same group, and the variations by each staff member are analyzed by statistically processing the number of responses by each staff member 72.

[0123] (Step S503) The analysis unit 112 aggregates the assessment results and determines the stage according to the likelihood of leaving. FIG. 21 is a schematic diagram illustrating the stages used to assess the likelihood of leaving. In this embodiment, for example, the stage is assessed on a five-level scale, from stages 1 to 5. Stage 1 is assessed as the lowest likelihood of leaving, and stage 5 is assessed as the highest. The analysis unit 121 aggregates the number of NG assessments for items in each group of element 1, element 2, and phase shown in FIGS. 19 and 20, and calculates an aggregate value. This aggregate value may be calculated by simply counting, or may be weighted (multiplied by a coefficient) according to each item number and then summed. For example, the weighting coefficients for indicators i5 and i11 related to room visits are set to be larger than the weighting coefficients for the other indicators i1 to i4, etc. The analysis unit 112 multiplies the aggregate values ​​of element 1, element 2, and phase, and determines the stage from the resulting evaluation value. For example, evaluation values ​​below 100 are judged to be stage 1, numbers in the 100s are judged to be stage 2, and numbers in the 200s, 300s, and 400s and above are judged to be stages 3, 4, and 5, respectively.

[0124] Furthermore, after determining the stage based on the aggregated value, if there is a record of a complaint from a user of the nursing care facility 900 about the monitoring support system 30, the stage is raised by one rank. This record of complaint is, for example, a complaint made by the user to the service provider via email, telephone, etc., regarding the monitoring support service. It is entered by the employee 76 via the terminal device 20 and stored in the memory unit 130. With the above processing, the subroutine processing of FIG. 17 is completed, and the process returns to the processing of FIG. 16.

[0125] (Step S45) The output unit 113 outputs the analysis result of the withdrawal possibility. For example, the output unit 113 transmits the analysis result to the terminal device 20.

[0126] The employee 76 displays a plurality of analysis results of the likelihood of withdrawal for each nursing care facility 900 side by side on a dashboard displayed on the display unit 230 of the terminal device 20. In this case, the output unit 113 outputs the analysis results for each nursing care facility 900. The analysis results include the results of which stage each nursing care facility 900 is in. For example, a service provider (such as the employee 76) can understand the likelihood of withdrawal from the monitoring support service at a specific nursing care facility 900 by referring to the dashboard. The service provider can compare each nursing care facility 900 by viewing the analysis results output for each nursing care facility 900. For example, even if the number of nursing care facilities 900 to which a service provider (equipment manager) provides services increases, the analysis results for each nursing care facility 900 can be easily compared, which has the effect of enabling early response to dissatisfaction among users of the equipment or service.

[0127] Furthermore, the analysis unit 112 may output an analysis result indicating which stage (see FIG. 21 ) the user is currently in. In this case, if the stage is above a predetermined level, i.e., if there is a high possibility of withdrawal from the monitoring support service of the monitoring support system 30, the analysis result may include an alert. The output unit 113 outputs this alert along with the analysis result. For example, the alert may be output as a report containing response actions. The report may include a report indicating the poor utilization of the monitoring support service based on the relevant factors (factor 1, factor 2, phase) and the NG indicators. When multiple response actions are output based on multiple NG indicators, the output may include the priority of the response actions. For example, a priority for each indicator may be predetermined, and the priority of the response action corresponding to the indicator with a higher priority may be higher. For example, if indicators i5 and i11 related to room visits are NG, the priority of the response action for these indicators may be higher than the priority of other response actions. This allows the service provider to quickly address complaints from users of the device or service. The service provider can read that the utilization of the monitoring support system has decreased and that some kind of trouble has occurred through the dashboard displayed on the display unit 230. Then, the service provider can grasp the possibility that the nursing care facility 900 will drop out of the monitoring support service and take measures to prevent the dropout before it happens, depending on the cause.

[0128] As described above, the information processing device according to the present embodiment includes an acquisition unit that acquires operation log data from a monitoring support system that detects the status of residents in each of multiple rooms in a nursing care facility using sensors installed in each of the rooms, an analysis unit that analyzes the likelihood of users of the monitoring support system dropping out of the service based on the operation log data, and an output unit that outputs the analysis results. This makes it possible to provide information that is useful for identifying dissatisfaction among device users in advance, i.e., information that is useful for identifying signs of dropping out due to dissatisfaction with the user's monitoring support service in advance, and ultimately makes it possible to prevent users from dropping out of the monitoring support service.

[0129] (Second embodiment) Next, an information processing device 10b according to a second embodiment will be described with reference to Figs. 22 to 24. The configuration other than that shown in Figs. 22 and 3A to 16 is the same as that of the information processing device 10 according to the first embodiment, and therefore description thereof will be omitted. Fig. 22 is a block diagram showing a schematic configuration of the information processing device 10b according to the second embodiment. Fig. 23 is a subroutine flowchart showing the processing of step S44 in Fig. 16 performed in the second embodiment. Fig. 24 is a flowchart showing a machine learning method for a trained model.

[0130] 22, the information processing device 10b includes a control unit 110, a communication unit 120, and a storage unit 130, similar to the first embodiment. The control unit 110 includes a learning unit 115 in addition to an acquisition unit 111, an analysis unit 112, and an output unit 113. The learning unit 115 creates and updates a trained model by the machine learning method shown in FIG. 24. The trained model is stored in the storage unit 130.

[0131] (Step S511) 23, the analysis unit 112 refers to the operation log data and the response information log data, and calculates a plurality of indicators of withdrawal signs by evaluating the operating status or utilization status of the watching support system 30 or the watching support service. This process is the same as step S501 in FIG. 17 in the first embodiment.

[0132] (Step S512) The analysis unit 112 analyzes the churn possibility using a trained model trained by the learning device 115, which is stored in the storage unit 130. Specifically, by inputting a plurality of indices (indices i1 to i12), a churn score indicating the degree of churn possibility is obtained as an output. Furthermore, as a subsequent process, a stage (see FIG. 21) may be determined according to the churn score by rule-based processing. After that, the process returns to the process of FIG. 16, and the output process of step S45 is performed.

[0133] (machine learning methods) 24 is a flowchart showing a machine learning method for a trained model. Below, a learning method using a neural network (NN) configured by combining perceptrons in a learning device will be described, but the present invention is not limited to this, and various supervised learning methods can be used. For example, random forests, support vector machines (SVMs), boosting, Bayesian network linear discriminant analysis, nonlinear discriminant analysis, etc. can be applied.

[0134] (Step S61) The learning device 115 reads learning sample data, which is training data. If it is the first time, the first set of learning sample data is read, and if it is the nth time, the nth set of learning sample data is read. The learning sample data is a set of multiple indicators (e.g., indicators i1 to i12) of the nursing care facility 900 and withdrawal results.

[0135] (Step S62) The learning device inputs the input data from the read learning sample data into the neural network.

[0136] (Step S63) The learner compares the neural network's estimation results, i.e., the estimated churn results, with the training data.

[0137] (Step S64) The learning device adjusts and updates the parameters based on the comparison results, for example, by performing a process called back-propagation (back-propagation) to reduce the error in the comparison results.

[0138] (Step S65) If the learning device has completed processing of all data from the 1st to nth sets (YES), the process proceeds to step S66; if not (NO), the process returns to step S61, the next learning sample data is read, and the process from step S61 onwards is repeated.

[0139] (Step S66) The learning device stores the trained model constructed in the processing up to this point in the storage unit 130 and then ends the processing (END). In the processing of FIG. 23 described above, the trained model generated in this way is used to estimate the likelihood of departure.

[0140] By analyzing the likelihood of churn using a machine-learned model in this way, as in the first embodiment, it becomes possible to identify signs of churn due to user dissatisfaction with the monitoring support service in advance and prevent churn from occurring.

[0141] The configurations of the information processing devices 10, 10b and the information processing system 1 including the same described above have been described as the main components in explaining the features of the above-described embodiments. However, the present invention is not limited to the above-described configurations and can be modified in various ways within the scope of the claims. For example, the information processing system 1 may be configured with the information processing device 10 and a terminal device 20 (display device) including a display unit 23. Furthermore, the processing in the information processing device 10 according to the above-described embodiments may include steps other than those shown in the flowcharts and sequence charts, or may not include some of the steps described above. Furthermore, the order of the steps is not limited to the above-described embodiments. Furthermore, each step may be combined with other steps and executed as a single step, may be included in other steps and executed, or may be divided into multiple steps and executed.

[0142] Furthermore, the information processing device 10 acquires the operation log data and the corresponding information log data, but may acquire only the operation log data and analyze the possibility of withdrawal based on this operation log data.

[0143] Furthermore, the means and methods for performing various processes in the information processing device or information processing system according to the above-described embodiments can be realized by either a dedicated hardware circuit or a programmed computer. The program may be provided, for example, by a computer-readable recording medium such as a USB memory or a DVD (Digital Versatile Disc)-ROM, or may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is typically transferred and stored in a storage unit such as a hard disk. The program may also be provided as standalone application software, or may be incorporated as a function into the software of a device such as a detection unit. [Explanation of symbols]

[0144] 1. Information Processing Systems 10, 10b Information processing device 110 control section 111 Acquisition Department 112 Analysis Department 113 Output section 115 Learning Machine 120 Communications Department 130 Storage section 20 Terminal equipment 30 Monitoring support system 31 Detection unit 311 Control Unit 31a Event Judgment Section 312 Communications Department 313 Camera 314 Care Call Department 315 Audio input / output section 316 Pause Switch 32 Monitoring device 321 Control Unit 32a Notification section 32b Notification settings section 322 Communications Department 323 Storage section 33 Terminal Equipment 34 Staff terminal 341 Control Unit 342 Radio Communication Department 343 Display section 344 Input section 345 Audio Input / Output Unit 70 Residents 72 staff 74 Administrator 76 employees

Claims

1. an acquisition unit that acquires operation log data of devices present in a nursing care facility; an analysis unit that analyzes the likelihood of a user using the device leaving the device based on the operation log data; an output unit that outputs the analysis result by the analysis unit; Equipped with The acquisition unit acquires staff response information log data via the staff terminal, the analysis unit calculates a plurality of churn probability indicators by analyzing the operation log data and the response information log data for a predetermined period of time; An information processing device, wherein the indicator of the likelihood of withdrawal includes an indicator that evaluates the operation or usage status of the device.

2. The information processing device described in Claim 1, wherein the analysis unit determines, for each of the indicators, whether the indicator is outside a specified range or whether the rate of change of the indicator per specified period is outside a specified range, and analyzes the possibility of withdrawal based on a stage determined according to the number of indicators determined to be outside the specified range.

3. The information processing device according to claim 1 , wherein the analysis unit estimates the likelihood of churn using a trained model trained with training sample data that combines a plurality of the indicators and churn results.

4. The information processing device according to claim 1 , wherein the output unit outputs a response action for the withdrawal in the analysis result.

5. The information processing device according to claim 4 , wherein the response actions are plural, and the output unit includes priorities of the response actions in the output.

6. The acquisition unit acquires the operation log data from each of the plurality of nursing care facilities, The analysis unit analyzes the possibility of withdrawal in each of the plurality of nursing care facilities, The information processing device according to claim 1 , wherein the output unit outputs the analysis results for each of the plurality of nursing care facilities for each of the nursing care facilities.

7. a display device having a display unit that displays the analysis results; An information processing device according to any one of claims 1 to 6; An information processing system comprising:

8. Furthermore, the equipment present in nursing care facilities, The information processing system according to claim 7 .

9. A churn likelihood analysis method executed by an information processing device, comprising: Step (a) of acquiring operation log data of devices present in a nursing care facility; (b) analyzing the likelihood of users using the device dropping out based on the operation log data; Step (c) of outputting the analysis results of step (b); Including, In the step (a), log data of staff response information is acquired through the staff terminal; In the step (b), the operation log data and the response information log data for a predetermined period are analyzed to calculate a plurality of churn probability indicators; A churn likelihood analysis method that executes processing in which the churn likelihood indicator includes an indicator that evaluates the usage status regarding the operation or utilization of the device.

10. A method for analyzing the possibility of withdrawal as described in claim 9, wherein in step (b), for each of the indicators, it is determined whether the indicator is outside a predetermined range or whether the rate of change of the indicator per predetermined period is outside a predetermined range, and the possibility of withdrawal is analyzed based on a stage determined according to the number of indicators determined to be outside the predetermined range.

11. 10. The churn likelihood analysis method according to claim 9, wherein in the step (b), the churn likelihood is estimated using a trained model trained with training sample data that combines a plurality of the indicators and churn results.

12. 12. The churn likelihood analysis method according to claim 9, wherein in the step (c), a corresponding action for churn in the analysis result is output.

13. The method for analyzing the likelihood of churn according to claim 12 , wherein the response actions are plural, and in step (c), priorities of the response actions are included in the output.

14. In the step (a), the operation log data is acquired from each of the plurality of nursing care facilities; In the step (b), the possibility of withdrawal is analyzed in each of the plurality of nursing care facilities; The withdrawal probability analysis method according to any one of claims 9 to 13, wherein in step (c), analysis results for each of the plurality of nursing care facilities are output for each of the nursing care facilities.

15. A control program for causing a computer to execute the churn likelihood analysis method according to any one of claims 9 to 14.

Images