JP7345323B2 - Cleaning load management device, cleaning load management system, cleaning load management method and program - Google Patents

Description

The present invention relates to a cleaning load management device, a cleaning load management system, a cleaning load management method, and a program.

BACKGROUND ART In accommodation facilities such as hotels and inns, it is necessary for cleaning staff to efficiently clean guest rooms during times when guests are not in their guest rooms. In contrast, there are multiple data tables that manage the working status of cleaning staff who clean multiple rooms, the check-out time of guests staying in each room, the cleaning status of each room, and the cleaning work time for each room. and a keyboard for instructing the creation of a cleaning plan for the accommodation facility, and when an instruction to create a cleaning plan is given, the data table is referenced and each cleaning staff on duty selects the first room to be cleaned within the scope of their responsibility and the cleaning start time. At the same time, the system calculates point values for determining the cleaning order for all rooms that are not occupied by guests and that have not yet been cleaned, and based on the calculated point values, the cleaning plan for each room is determined by each cleaning staff. An accommodation facility cleaning plan creation system has been proposed that includes a hard disk drive that stores a cleaning plan program to be created (for example, see Patent Document 1).

Japanese Patent Application Publication No. 2003-67509

However, the accommodation facility cleaning plan creation system described in Patent Document 1 does not take into account that the degree of dirt in each room changes depending on how the guests spend time in each room. If the room is dirty, it will take time to clean the room, and there is a risk that the cleaning will not be carried out according to the created cleaning plan.

The present invention has been made in view of the above reasons, and provides a cleaning load management device, a cleaning load management system, a cleaning load management method, and a program that can contribute to the creation of an appropriate cleaning plan that takes into account the degree of dirtiness of a room. The purpose is to

In order to achieve the above object, the cleaning load management device according to the present invention includes:
an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in the room;
a cleaning load acquisition unit that acquires load performance information indicating a cleaning load that is a workload required for cleaning the room;
a coefficient determination unit that determines a neural network coefficient of a neural network for determining a cleaning load when cleaning the room after a user stays in the room, based on the operation history information and the load performance information; ,
Using the neural network for which the neural network coefficients have been determined by the coefficient determining unit, the neural network of the room is determined based on operation history information during a cleaning load prediction target period that includes at least a portion of the period during which the user stayed in the room. A neural network calculation unit that calculates a cleaning load when cleaning is performed.

According to the present invention, the neural network calculation unit uses the neural network whose neural network coefficients have been determined by the coefficient determination unit to calculate the cleaning load when cleaning a room from the operation history information in the cleaning load prediction target period. demand. This makes it possible to estimate the cleaning load based on the usage status of the user's room and how dirty the room is after the user has stayed in the room. The optimal cleaning work time can be calculated and reflected in the cleaning worker's cleaning schedule. Therefore, wasted time in the cleaning schedule can be reduced, and the efficiency of cleaning work can be improved.

Schematic configuration diagram of a cleaning load management system according to Embodiment 1 of the present invention A block diagram showing the hardware configuration of an operating device and an air conditioner according to Embodiment 1. Functional block diagram of operating equipment and air conditioner according to Embodiment 1 Block diagram showing the hardware configuration of a terminal device according to Embodiment 1 Functional block diagram of a terminal device according to Embodiment 1 Block diagram showing the hardware configuration of the cleaning load management device according to the first embodiment Functional block diagram of the cleaning load management device according to the first embodiment A diagram illustrating an example of information stored by the operation history storage unit according to Embodiment 1. A diagram showing an example of information stored in the cleaning load storage unit according to Embodiment 1. A diagram for explaining an example of the relationship between load performance information and operation history according to Embodiment 1 A diagram showing an example of information stored in the guest room information storage unit according to Embodiment 1. Explanatory diagram of a neural network used in the cleaning load management device according to Embodiment 1 A diagram showing an example of information stored in the worker information storage unit according to Embodiment 1. (A) is an explanatory diagram of the operation when determining the weighting coefficient of the neural network of the cleaning load management device according to the first embodiment, and (B) is a cleaning diagram in which the cleaning load management device according to the first embodiment shows the predicted cleaning load. Diagram explaining the operation when outputting load information A sequence diagram showing an example of the operation of the cleaning load management system according to the first embodiment A diagram showing an example of an operation screen image displayed on the display unit of the terminal device according to Embodiment 1. A sequence diagram showing an example of the operation of the cleaning load management system according to the first embodiment (A) is a diagram showing an example of an operation screen image displayed on the display unit of the terminal device according to Embodiment 1, and (B) is a diagram showing an example of an operation screen image displayed on the display unit of the terminal device according to Embodiment 1. Diagram showing another example of the screen image Flowchart showing an example of the flow of operation history information/environmental performance information management processing executed by the operating device according to Embodiment 1 Flowchart showing an example of the flow of terminal control processing executed by the operating device according to Embodiment 1 Flowchart showing an example of the flow of cleaning load management processing executed by the cleaning load management device according to Embodiment 1 Flowchart showing an example of the flow of coefficient determination processing executed by the cleaning load management device according to Embodiment 1 Schematic configuration diagram of a cleaning load management system according to Embodiment 2 of the present invention Functional block diagram of cleaning load management device according to Embodiment 2 Block diagram showing the hardware configuration of a cloud server according to Embodiment 2 Functional block diagram of cloud server according to Embodiment 2 Sequence diagram showing an example of the operation of the cleaning load management system according to Embodiment 2 Sequence diagram showing an example of the operation of the cleaning load management system according to Embodiment 2 Sequence diagram showing an example of the operation of the cleaning load management system according to Embodiment 2 Flowchart showing an example of the flow of cleaning load management processing executed by the cleaning load management device according to Embodiment 2 Flowchart showing an example of the flow of coefficient information generation processing executed by the cloud server according to Embodiment 2 A diagram showing an example of an operation screen image displayed on a display unit of a terminal device according to a modification example.

(Embodiment 1)
Hereinafter, a cleaning load management device according to an embodiment of the present invention will be described with reference to the drawings. The cleaning load management device according to the present embodiment uses a neural network to calculate the cleaning load of the cleaning worker when cleaning the room based on the operation history of each of at least one device installed in the room during the target period. This is what we seek. This cleaning load management device includes an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in a room, and a load record indicating the cleaning load that is the work load required to clean the room. It includes a cleaning load acquisition section that acquires information, a coefficient determination section, and a neural network calculation section. Here, the coefficient determining unit uses a neural network to determine a cleaning load when cleaning a room after the user has stayed for a preset target period, based on the acquired operation history information and load performance information. Determine the neural network coefficients of the network. Further, the neural network calculation unit uses the neural network whose neural network coefficients have been determined by the coefficient determination unit to determine the cleaning load when cleaning the room from the operation history information in the target period.

For example, as shown in FIG. 1, the cleaning load management system according to the present embodiment includes a cleaning load management device 1 installed in an accommodation facility H, and a terminal device 2 owned by a cleaning worker who cleans the accommodation facility. and an operating device 4 installed in each guest room RA, RB in the accommodation facility H. The cleaning load management system also includes a customer server 10 that can communicate with the cleaning load management device 1 via the wide area network NW1. The wide area network NW1 is, for example, the Internet. Further, each guest room RA, RB of the accommodation facility H is equipped with an air conditioner 5, a lighting fixture 6, and a bathroom ventilation fan 7. The user can operate the air conditioner 5, the lighting equipment 6, and the bathroom ventilation fan 7 using the operating device 4. Here, the operating device 4 and the air conditioner 5 may be connected, for example, by wire, and the operating device 4, the lighting fixture 6, and the bathroom ventilation fan 7 may be connected, for example, wirelessly. Further, the operating device 4 may be able to be wirelessly connected to, for example, a mobile terminal (not shown) owned by a user of the guest rooms RA and RB. In this case, when the user operates the mobile terminal, operation information indicating the operation details for the air conditioner 5, lighting fixture 6, or bathroom ventilation fan 7 is transmitted from the mobile terminal to the operating device 4, and the operating device 4 Control information corresponding to the operation information received from the controller may be transmitted to the air conditioner 5, the lighting equipment 6, or the bathroom ventilation fan 7.

The cleaning load management device and each air conditioner 5 are connected to a local network NW2 installed within the accommodation facility H. The local network NW2 is, for example, a wired or wireless LAN (Local Area Network) that conforms to a communication standard such as Ethernet (registered trademark). Furthermore, the local network NW2 is connected to the wide area network NW1 via a router 91 and a data line termination device 92. The router 91 is, for example, a broadband router, and the data line terminal device 92 has functions such as a model and a gateway. Furthermore, a wireless communication device 8 installed in the accommodation facility H is connected to the local network NW2, and the terminal device 2 can communicate with the cleaning load management device 1 via the wireless communication device 8. . The wireless communication device 8 functions, for example, as a wireless LAN access point, and performs communication with the terminal device 2 using a communication method that conforms to the wireless LAN standard of, for example, IEEE802.11ac.

The customer server 10 is a server equipped with a large-capacity storage, and manages information indicating the reservation status of each room user in the accommodation facility H, information indicating the stay status, and the like.

For example, as shown in FIG. 2, the operating device 4 includes a CPU (Central Processing Unit) 401, a main storage section 402, an auxiliary storage section 403, a display section 404, an input section 405, a device communication interface 407, A bus 409 that connects these to each other is provided. The main storage unit 402 is a volatile memory such as RAM (Random Access Memory), and is used as a work area for the CPU 401. The auxiliary storage unit 403 is a nonvolatile memory such as a semiconductor flash memory, and stores programs for the CPU 401 to execute various processes. The display unit 404 is, for example, a display device such as a liquid crystal display. The input unit 405 is a touch pad disposed overlapping the display unit 404, a button switch disposed on a part of the casing of the operating device 4, or the like. The device communication interface 407 is for communicating with the air conditioner 5, and is, for example, a wired communication interface. For example, as shown in FIG. 2, the air conditioner 5 includes a control unit 510 for controlling the operation of a compressor (not shown) or a blower fan (not shown). The air conditioner 5 also includes a temperature sensor (not shown) that detects the temperature of the guest room RA (RB) in which the air conditioner 5 is installed, and a temperature sensor (not shown) that detects the temperature of the guest room RA (RB) in which the air conditioner 5 is installed. A humidity sensor (not shown) for detection is provided. The control unit 510 includes a CPU 501, a main storage unit 502, an auxiliary storage unit 503, a local communication interface 506, a device communication interface 507, and a bus 509 that connects these to each other. The main storage unit 502 is used as a work area for the CPU 501, and the auxiliary storage unit 503 stores programs for the CPU 501 to execute various processes. The local communication interface 506 is, for example, a wired communication interface or a wireless module, and is connected to the local network NW2. The device communication interface 507 is for communicating with the operating device 4, and is, for example, a wired communication interface.

By reading the program stored in the auxiliary storage unit 403 into the main storage unit 402 and executing it, the CPU 401 executes the program, for example, as shown in FIG. 414 , an operation history acquisition section 415 , an operation history transmission section 416 , an environmental information acquisition section 417 , an environmental performance transmission section 418 , and a time measurement section 419 . The auxiliary storage unit 403 shown in FIG. 2 also includes a control information storage unit 431 that stores control information generated by the control information generation unit 412, an operation history storage unit 432, and a display unit 404, as shown in FIG. It has an image storage unit 433 that stores image information indicating an operation screen image to be displayed on the screen, and an environmental performance storage unit 434. The operation history storage unit 432 is a combination of control information transmitted from the operating device 4 to the air conditioner 5, lighting fixture 6, and bathroom ventilation fan 7, and date and time information indicating the date and time when the control information was transmitted. Stores operation history information. The environmental performance storage unit 434 shows environmental information obtained from the air conditioner 5 indicating environmental parameters such as temperature and humidity of the guest room RA (RB) in which the air conditioner 5 is installed, and the date and time when the environmental information was obtained. Stores environmental performance information that is a combination of date and time information.

The reception unit 411 receives operation information indicating the content of the operation performed on the input unit 405 by the user of the guest room RA (RB). The control information generation unit 412 generates control information for the air conditioner 5, lighting fixture 6, or bathroom ventilation fan 7 corresponding to the operation information received by the reception unit 411, and stores the generated control information in the control information storage unit 431. . The display control unit 413 acquires image information corresponding to the operation information received by the reception unit 411 from the image storage unit 433, and causes the display unit 404 to display an operation screen image corresponding to the acquired image information. The control information transmitting unit 414 transmits the control information stored in the control information storage unit 431 to the air conditioner 5, the lighting fixture 6, or the bathroom ventilation fan 7. Here, the control information transmitting section 414 transmits the updated control information to the air conditioner 5, the lighting fixture 6, or the bathroom ventilation fan 7, for example, when the control information stored in the control information storage section 431 is updated. When the control information transmitting unit 414 transmits the control information to the air conditioner 5, the lighting fixture 6, or the bathroom ventilation fan 7, the operation history acquisition unit 415 stores the transmitted control information as a date and time indicating the date and time when the control information was transmitted. It is stored in the operation history storage unit 432 in association with the information.

When the operation history transmitter 416 receives teacher information request information, which will be described later, from the cleaning load management device 1 via the air conditioner 5, the operation history transmitter 416 sends the information within the past preset teacher information target period stored in the operation history storage section 432. All of the operation history information included in is transmitted to the cleaning load management device 1 via the air conditioner 5. Here, the teacher information target period is set, for example, to a period from a time point that goes back a period of one week to one year from the most recent time point in the guest room RA (RB) to the most recent time point. For example, assume that the teacher information target period is set to a period from one year back to the most recent point in time. In this case, when the operation history storage section 432 only stores operation history information from a point one week back to the most recent point in time, the operation history transmission section 416 transmits the operation history information to the operation history storage section 432. Sends all the operation history information stored by.

Further, upon receiving operation history request information, which will be described later, from the cleaning load management device 1 via the air conditioner 5, the operation history transmitting unit 416 transmits the information within a preset load determination target period stored in the operation history storage unit 432. The operation history information is transmitted to the cleaning load management device 1 via the air conditioner 5. Here, the load determination target period is set, for example, to a period from when the user who most recently stayed in guest room RA (RB) checks in until he checks out.

The environmental information acquisition unit 417 sends environmental request information that requests the air conditioner 5 to transmit environmental information indicating environmental parameters such as temperature and humidity of the guest room RA (RB) in which the air conditioner 5 is installed. The environmental information is acquired from the air conditioner 5 by transmitting it to the air conditioner 5. Then, the environmental information acquisition unit 417 causes the environmental performance storage unit 434 to store a combination of the acquired environmental information and date and time information indicating the date and time when the environmental information was acquired as environmental performance information. Upon receiving teacher information request information from the cleaning load management device 1 via the air conditioner 5, the environmental performance transmitting unit 418 transmits the environmental performance information included in the above-mentioned teacher information target period stored in the environmental performance storage unit 434. are transmitted to the cleaning load management device 1 via the air conditioner 5. For example, assume that the teacher information target period is set to the period from one year back in the past to the most recent point in time, as described above. In this case, when the environmental performance storage unit 434 only stores operation history information from a point one week back to the most recent time, the environmental performance transmission unit 418 transmits the information to the environmental performance storage unit 434 Sends all the operation history information stored by. Further, upon receiving environmental performance request information to be described later from the cleaning load management device 1 via the air conditioner 5, the environmental performance transmitting unit 418 transmits the environmental Performance information is transmitted to the cleaning load management device 1 via the air conditioner 5.

The timer unit 419 determines whether the performance acquisition time for acquiring the operation history information and the environmental performance information has arrived, based on time information output from a software timer (not shown). The performance acquisition timing is set to arrive every 10 minutes, for example. When the performance acquisition time arrives, the clock unit 419 outputs performance acquisition time notification information to the operation history acquisition unit 415 and the environmental information acquisition unit 417.

Returning to FIG. 2, the CPU 501 of the air conditioner 5 reads out the program stored in the auxiliary storage unit 503 to the main storage unit 502 and executes it, thereby controlling the operation control unit 511, environmental information, etc., as shown in FIG. 3, for example. It functions as a generation section 512 and a relay section 513. The operation control unit 511 controls the operation of the compressor and the blower fan based on control information received from the operating device 4. The environmental information generating unit 512 generates environmental information indicating environmental parameters such as the temperature detected by the temperature sensor and the humidity detected by the humidity sensor, and transmits the generated environmental information to the operating device 4. Upon receiving the teacher information request information, operation history request information, and environmental performance request information from the cleaning load management device 1, the relay unit 513 transmits the received teacher information request information, operation history request information, and environmental performance request information to the operating device 4. Send to. Further, upon receiving the operation history information and environmental performance information from the operating device 4, the relay unit 513 transmits the received operation history information and environmental performance information to the cleaning load management device 1.

The terminal device 2 is, for example, a mobile information terminal, and as shown in FIG. 4, includes a CPU 201, a main storage section 202, an auxiliary storage section 203, a display section 204, an input section 205, a wireless module 206, A bus 209 that connects these to each other is provided. The main storage unit 202 is used as a work area for the CPU 201, and the auxiliary storage unit 203 stores programs for the CPU 201 to execute various processes. The display unit 204 is, for example, a display device such as a liquid crystal display. The input unit 205 is a touch pad placed over the display unit 204 . The wireless module 206 is for communicating with the wireless communication device 8, and performs communication using a communication method that conforms to the wireless LAN communication standard of IEEE802.11ac, for example.

By reading the program stored in the auxiliary storage unit 203 into the main storage unit 202 and executing it, the CPU 201 executes the program, for example, as shown in FIG. 214 and a schedule acquisition section 215. Further, as shown in FIG. 5, the auxiliary storage unit 203 shown in FIG. and an image storage unit 233 that stores image information indicating an operation screen image to be displayed on the screen 204 . The load record storage unit 232 stores load record information indicating the record of the cleaning load of the cleaning worker who owns the terminal device 2 .

The reception unit 211 receives operation information indicating the content of the operation performed by the cleaning worker on the input unit 205. The display control unit 212 acquires image information corresponding to the operation information received by the reception unit 211 from the image storage unit 233 and causes the display unit 204 to display an operation screen image corresponding to the acquired image information. The load performance acquisition unit 213 acquires load performance information indicating the performance of the cleaning load based on the operation information received by the reception unit 211, and stores the acquired load performance information in the load performance storage unit 232. The load record transmitting unit 214 transmits the load record information stored in the load record storage unit 232 to the cleaning load management device 1 according to the operation information received by the receiving unit 211. The schedule acquisition unit 215 acquires schedule information from the cleaning load management device 1 by transmitting schedule request information requesting the cleaning load management device 1 to transmit schedule information according to the operation information received by the reception unit 211. do. Then, the schedule acquisition unit 215 causes the schedule storage unit 231 to store the acquired schedule information. At this time, the display control unit 212 causes the display unit 204 to display an operation screen image for notifying the cleaning worker of the cleaning schedule indicated by the schedule information stored in the schedule storage unit 231.

For example, as shown in FIG. 6, the cleaning load management device 1 includes a CPU 101, a main storage section 102, an auxiliary storage section 103, a local communication interface 104, and a bus 109 that connects these to each other. CPU 101 is, for example, a multi-core processor. The main storage unit 102 is used as a work area for the CPU 101, and the auxiliary storage unit 103 stores programs for the CPU 101 to execute various processes. The local communication interface 104 is, for example, a wired communication interface or a wireless module, and is connected to the local network NW2.

By reading the program stored in the auxiliary storage unit 103 into the main storage unit 102 and executing it, the CPU 101 reads out the operation history acquisition unit 111, the cleaning load acquisition unit 112, the guest room information acquisition unit 113, and the environment, as shown in FIG. It functions as a performance acquisition section 119, a neural network calculation section 114, a coefficient determination section 115, a coefficient setting section 116, a schedule generation section 117, a schedule transmission section 118, and a time measurement section 120. Further, the auxiliary storage unit 103 shown in FIG. 6 includes an operation history storage unit 131, a cleaning load storage unit 132, a guest room information storage unit 133, an environmental performance storage unit 138, and a neural network storage unit, as shown in FIG. 134, a predicted load storage section 135, a schedule storage section 136, and a worker information storage section 137. For example, as shown in FIG. 8, the operation history storage unit 131 stores information indicating the operating states of the air conditioner 5, the lighting fixture 6, and the bathroom ventilation fan 7 in association with date and time information. In the example shown in FIG. 8, the operation history storage unit 131 stores, for the air conditioner 5, operation information indicating the on/off state, operation mode information indicating the operation mode such as cooling or ventilation, and air volume information indicating the set air volume. Set temperature information indicating the set temperature is stored in association with date and time information. The operation history storage unit 131 also stores operation information indicating the on/off state of the lighting fixtures 6 installed in each of the bedroom and bathroom of the guest room RA (RB) in association with date and time information. Further, the operation history storage unit 131 stores operation information indicating the on/off state of the bathroom ventilation fan 7 and air volume information indicating the set air volume in association with date and time information.

In the case of the example shown in FIG. 8, after the user checks in on May 28, at 17:00, the air conditioner 5 installed in the bedroom is turned on with the operation mode "cooling", air volume "strong", and the set temperature " 25° C., and the lighting equipment 6 is turned on. Next, at 20:00 on May 28th, the user changes the operation mode of the air conditioner 5 in the bedroom to "ventilation" and turns on the bathroom ventilation fan 7 installed in the bathroom with the air volume set to "low". At the same time, the lighting equipment 6 installed in the bathroom is turned on. Thereafter, at 0:00 on May 29th, the user turns off the air conditioner 5 and lighting fixture 6 installed in the bedroom, and also turns off the bathroom ventilation fan 7. Next, at 6:00 on May 29th, the user turned on the air conditioner 5 installed in the bedroom with the operation mode set to "cooling," the air flow rate to "low," and the set temperature to "27°C." The lighting equipment 6 installed in the bedroom is turned on, and the bathroom ventilation fan 7 is turned on with the air volume set to "strong". Subsequently, at 6:10 on May 29th, the user changed the settings of the air conditioner 5 installed in the bedroom to the operating mode "air purification", the air volume "strong", and the set temperature "25 degrees Celsius". ing. Thereafter, at 7:35 a.m. before checkout on May 29, the user changed the operating mode of the air conditioner 5 installed in the bedroom to "ventilation", turned off the lighting equipment installed in the bedroom, and The air volume of the bathroom ventilation fan 7 has been changed to "low".

For example, as shown in FIG. 9, the cleaning load storage unit 132 stores load performance information including work time index information representing an index of the length of cleaning work time and workload index information representing an index of the load of cleaning work. , the room identification information and work date information indicating the day on which the cleaning work was performed are stored in association with each other. In the example shown in FIG. 9, the work time index information indicates a numerical value when the length of the work time required for the cleaning work is expressed in five stages, and the smaller the value is, the longer the work time is. Further, the workload index information indicates a numerical value when the load applied to the cleaning worker during cleaning work is expressed in five stages, and the smaller the numerical value is, the greater the workload is. Here, the relationship between the working time index and the workload index included in the load performance information and the operation history of each of the air conditioner 5, lighting fixture 6, and bathroom ventilation fan 7 in the guest room RA (RB) will be explained. For example, as shown in FIG. 10, the deodorizing air purification button of the air conditioner 5 is used, the ventilation function of the air conditioner 5 is used with the air volume "low", the number of operations of the air conditioner 5 is small, and the bathroom ventilation fan 7 is turned on. When the air volume is set to "low", the lighting equipment 6 is turned on for a short time, the number of users of the guest room RA (RB) is small, and the number of children included in the users is small, the work time index and the work load index are It tends to get bigger. On the other hand, the deodorizing air purification button of the air conditioner 5 is not used, the ventilation function of the air conditioner 5 is used with the air volume set to "strong", the air conditioner 5 is operated many times, and the bathroom ventilation fan 7 is set to the air volume "strong". When the lighting equipment 6 is on for a long time, the number of users of the guest room RA (RB) is large, and the number of children included in the users is large, the working time index and the workload index tend to become smaller. .

For example, as shown in FIG. 11, the guest room information storage unit 133 stores room type information indicating the type of each of a plurality of rooms in the accommodation facility H, room area information indicating the size of the room, and users staying in the room. Room information including information on the number of guests staying in the room and the number of users staying in the room is stored. Further, the guest room information storage unit 133 stores guest room information in association with room identification information that identifies the room. Room type information includes a single room with a single bed, a double room with a double bed, a twin room with two single beds, a king room with two double beds, and a king room with three single beds. This information identifies each of the triple room and suite room where two double beds are installed and the bedroom and living room are partitioned. Further, the information on the number of guests staying includes information indicating "adults" and "children" as the types of users. In the example shown in FIG. 11, the room identification information, that is, the rooms with guest room numbers "301" and "302" are single rooms, and the rooms with guest room numbers "303" and "304" are double bedrooms, This indicates that the rooms with guest numbers "305" and "306" are twin rooms, and the rooms with guest numbers "501" and "502" are suite rooms. Each room is equipped with two double beds, and the guest room and living room are separated.

Returning to FIG. 7, the environmental performance storage unit 138 converts environmental performance information indicating past performance values of environmental parameters such as temperature and humidity in each room (for example, guest rooms RA and RB) in the accommodation facility H into date and time information. It is stored in correspondence.

Neural network storage unit 134 stores information regarding a neural network used by neural network calculation unit 114 and having a predetermined number of nodes and layers. Specifically, the neural network storage unit 134 stores information indicating the number of nodes and layers of the neural network, and the weighting coefficient and activation function corresponding to each node. This neural network is used to determine the cleaning load when cleaning guest room RA (RB) after the user has stayed for a preset target period. This neural network has an input layer L10, a hidden layer L20, and an output layer L30, as shown in FIG. The input layer L10 inputs numerical values indicated by each of the operation history information, environmental performance information, and guest room information to the hidden layer L20.

The hidden layer L20 includes a preset number M[j] of nodes x[j,i] (1≦i≦M[j], M[j] is a positive integer). ) consists of layers. That is, the hidden layer L20 has a structure in which each node string is connected. Here, the output y[j,i] of each node x[j,i] is expressed by the following relational expression (1).
Here, W[j, i, k] indicates a weighting coefficient, and f(*) indicates an activation function. This weighting coefficient W[j, i, k] corresponds to the neural network coefficient that determines the structure of the neural network described above. Further, as the activation function, a nonlinear function such as a sigmoid function, a ramp function, a step function, a softmax function, etc. is used. In the hidden layer L20, the information input to the nodes is the sum of the outputs of each node in the previous layer multiplied by a weighting coefficient. Then, the output of the activation function that takes the sum as an argument is transmitted to the next layer. The output layer L30 outputs the output y[j,i] from the final layer of the hidden layer L20 as is. The neural network storage unit 134 stores information indicating the structure of the neural network expressed by equation (1) and information indicating the weighting coefficient W[j, i, k] of the neural network. Further, the neural network storage unit 134 stores initial weighting coefficient information indicating initial values of weighting coefficients of the neural network.

Returning to FIG. 7, the predicted load storage unit 135 stores cleaning load information indicating the predicted value of the cleaning load calculated by the neural network calculation unit 114.

For example, as shown in FIG. 13, the worker information storage unit 137 stores information indicating the work history of each cleaning worker and work date information indicating the work date in association with worker identification information. The information indicating the work history includes period of service information indicating the period of service at the accommodation facility H where the worker currently works, and years of experience before work indicating the period of experience prior to working at the accommodation facility H where the worker currently works. include.

Returning to FIG. 7, the schedule storage unit 136 stores schedule information for each cleaning worker generated by the schedule generation unit 117.

The operation history acquisition unit 111 transmits operation history request information to the operating device 4 to acquire the operation history of the air conditioner 5, lighting equipment 6, and bathroom ventilation fan 7 installed in guest rooms RA and RB of the accommodation facility H. Obtain the operation history information shown. Here, the operation history request information is information for requesting the operating device 4 to transmit operation history information of each of the air conditioner 5, the lighting fixture 6, and the bathroom ventilation fan 7 during the cleaning load prediction period. The environmental performance acquisition unit 119 acquires environmental performance information indicating the environmental performance of the guest rooms RA and RB by transmitting environmental performance request information to the operating device 4. Here, the environmental performance request information is information for requesting the operating device to transmit the environmental performance information of the guest room RA (RB) in which the air conditioner 5, lighting equipment 6, and bathroom ventilation fan 7 are installed. . The guest room information acquisition unit 113 acquires guest room information from the customer server 10 by transmitting guest room information request information to the customer server 10 that requests the customer server 10 to transmit guest room information. The cleaning load acquisition unit 112 acquires load performance information indicating the cleaning load, which is the workload required for cleaning guest rooms RA and RB of the accommodation facility H, from the terminal device 2 .

The coefficient determination unit 115 determines the weighting coefficient of the neural network for determining the cleaning load described above, based on the operation history information, the environmental performance information, the guest room information, and the load performance information. The coefficient setting unit 116 first obtains initial weighting coefficient information indicating an initial weighting coefficient from the neural network storage unit 134, and sets the obtained initial weighting coefficient as a weighting coefficient of the neural network. Next, the coefficient determination unit 115 obtains the cleaning load calculated by the neural network calculation unit 114 using the neural network, as shown in FIG. 14(A), for example. Here, the neural network calculation unit 114 calculates the cleaning load using a neural network based on the operation history information DAT1, environmental performance information DAT2, and guest room information DAT3, which are teacher information. Subsequently, the coefficient determining unit 115 acquires the load performance information DAT4, which is teacher information stored in the cleaning load storage unit 132, and calculates the cleaning load indicated by the acquired load performance information DAT4 and the cleaning load calculated using the neural network. Calculate the error between Then, the coefficient determination unit 115 determines the weighting coefficient of the neural network by backpropagation based on the calculated error. Here, the coefficient determination unit 115 determines the weighting coefficient using, for example, an autoencoder. Then, the coefficient determining unit 115 causes the neural network storage unit 134 to store weighting coefficient information indicating the determined weighting coefficient.

The coefficient setting unit 116 sets the weighting coefficient determined by the coefficient determining unit 115 as the weighting coefficient of the neural network. The neural network calculation unit 114 uses the neural network whose weighting coefficient has been determined by the coefficient determination unit 115 to calculate the cleaning of the guest room RA (RB) from the operation history information, environmental performance information, and guest room information during the cleaning load prediction period. Find the cleaning load that indicates the expected cleaning load when performing the cleaning. Here, as shown in FIG. 14(B), the coefficient setting unit 116 uses the weighting coefficient determined by the coefficient determination unit 115 stored in the neural network storage unit 134 as the weight of the neural network used by the neural network calculation unit 114. Set to coefficient. Then, the neural network calculation unit 114 calculates the predicted cleaning load using the neural network in which the weighting coefficients have been set by the coefficient setting unit 116 from the operation history information DAM1, the environmental performance information DAC1, and the guest room information DAR1 in the cleaning load prediction target period. is calculated, and cleaning load information indicating the calculated predicted cleaning load is output. The neural network calculation unit 114 causes the predicted load storage unit 135 to store predicted load information indicating the cleaning load indicating the calculated predicted cleaning load.

Returning to FIG. 7, the schedule generation unit 117, based on the cleaning load calculated by the neural network calculation unit 114 stored in the predicted load storage unit 135 and the worker information stored in the worker information storage unit 137, Schedule information indicating a cleaning schedule for a cleaning worker is generated. Here, the schedule generation unit 117 calculates the cleaning work time to be allocated to the cleaning worker from the determined cleaning load using, for example, the relational expression (2) below.
(Cleaning work time) = (Cleaning load) x (C1 x (service period) + C2 x (experience period))
...Formula (2)
Here, C1 and C2 are preset coefficients. Then, the schedule generation unit 117 refers to the worker information storage unit 137 and selects a schedule for each guest room according to the period of service and period of experience of each cleaning worker who is working on the day when the cleaning work of guest rooms RA and RB is to be performed. Calculate the cleaning work time at RA and RB. Then, the schedule generation unit 117 generates the calculated cleaning work time for each guest room RA, RB, and usage status information indicating whether or not a user of a room (for example, guest room RA, RB) in the accommodation facility H is staying in the room. Generate schedule information based on and. The schedule generation unit 117 causes the schedule storage unit 136 to store the generated schedule information. The schedule transmitting unit 118 transmits the schedule information stored in the schedule storage unit 136 to the operating device 4.

The clock unit 120 determines whether the coefficient update time has arrived to update the weighting coefficients of the neural network stored in the neural network storage unit 134, based on time information output from a software timer (not shown). . The coefficient update time is set, for example, to arrive every day. When the coefficient update time arrives, the clock unit 120 outputs coefficient update time arrival notification information to the operation history acquisition unit 111, the environmental performance acquisition unit 119, the guest room information acquisition unit 113, and the coefficient determination unit 115.

Next, the operation of the cleaning load management system according to this embodiment will be described with reference to FIGS. 15 and 18. First, as shown in FIG. 15, when a preset performance information acquisition time arrives, the operating device 4 acquires control information from the control information storage section 431 and stores it in the operation history storage section 432 (step S1 ), the environmental information is acquired from the air conditioner 5 and stored in the environmental performance storage section 434 (step S2). Thereafter, the operating device 4 repeatedly executes the processes of steps S1 and S2 every time the performance information acquisition time arrives.

In addition, after cleaning the guest room RA (RB), the cleaning worker of the accommodation facility H cleans the guest room RA (RB), and with the operation screen image for inputting a questionnaire regarding the cleaning load displayed on the display unit 204 of the terminal device 2, the cleaning worker cleans the guest room RA (RB). Suppose that the user performs an operation to input information into the questionnaire via the input unit 205 of No. 2. Here, in the terminal device 2, an operation screen image GA1 for inputting a questionnaire, as shown in FIG. 16, for example, is displayed on the display unit 204. The operation screen image GA1 includes an area M1 that displays the guest room number and cleaning date of the cleaned guest room RA (RB), a message MQ1 that displays the content of the questionnaire regarding the cleaning time, and a message that displays the content of the questionnaire regarding the cleaning work load. and a message MQ2 indicating the message. Further, the operation screen image GA1 includes touch areas BA1 and BA2 for responding to the question content of each questionnaire using five-level indicators, and areas MA1 and MA2 for displaying the answer content. The cleaning worker then displays the operation screen image GA1 on the display unit 204 of the terminal device 2 after performing cleaning work on the guest room RA (RB), for example after the users of the guest room RA (RB) have checked out. In this state, the user answers the questionnaire by touching the portion corresponding to the index value to be answered displayed in the touch areas BA1 and BA2 on the input unit 205. Note that FIG. 16 shows a case where the guest room number RA (RB) in which the cleaning work was performed is "403" and the cleaning date is "May 29th." Furthermore, the message MQ1 is set to "Was the work time for cleaning the room long or short?", and the message MQ2 is set to "Was the cleaning work difficult?".

Returning to FIG. 15, when the cleaning worker performs an input operation to the questionnaire, the terminal device 2 receives operation information indicating the contents of the input operation to the questionnaire (step S3). Next, the terminal device 2 generates load record information indicating the cleaning load record corresponding to the operation information indicating the received operation details, and stores it in the load record storage unit 232 (step S4). Subsequently, when the terminal device 2 receives an operation for transmitting the questionnaire results after the cleaning worker inputs the questionnaire, the load record information stored in the load record storage unit 232 is stored in the terminal device 2. from there to the cleaning load management device 1 (step S5). On the other hand, upon receiving the load performance information, the cleaning load management device 1 causes the cleaning load storage unit 132 to store the received load performance information (step S6). Thereafter, each time the cleaning worker performs an input operation to the questionnaire, the series of processes from steps S3 to S6 described above are repeatedly executed.

After that, it is assumed that the time has come to update the preset weighting coefficients of the neural network used by the cleaning load management device 1. In this case, teacher information request information that requests the operating device 4 to transmit operation history information and environmental performance information as teacher information is transmitted from the cleaning load management device 1 to the operating device 4 (step S7). On the other hand, when the operating device 4 receives the teacher information request information from the cleaning load management device 1, operation history information and environmental performance information are transmitted from the operating device 4 to the cleaning load managing device 1 in response. S8). Here, the operation history information is operation history information of each of the air conditioner 5, lighting fixture 6, and bathroom ventilation fan 7 that are operated by the operation device 4, and the environmental performance information is the operation history information of the air conditioner 5, the lighting fixture 6, and the bathroom ventilation fan 7 that are operated by the operation device 4. This is environmental performance information showing the performance values of the temperature, humidity, etc. of the guest room RA (RB) in which the room RA (RB) is installed.

Next, teacher information request information requesting the customer server 10 to transmit guest room information as teacher information is transmitted from the cleaning load management device 1 to the customer server 10 (step S9). On the other hand, upon receiving the teacher information request information from the cleaning load management device 1, the customer server 10 specifies guest room information corresponding to the received teacher information request information (step S10). Subsequently, the identified guest room information is transmitted from the customer server 10 to the cleaning load management device 1 (step S11). On the other hand, the cleaning load management device 1 performs a coefficient determination process for determining the weighting coefficient of the neural network based on the received operation history information, environmental performance information, and guest room information, and the load performance information stored in the cleaning load storage unit 132. Execute (step S12). The cleaning load management device 1 causes the neural network storage unit 134 to store weighting coefficient information indicating the weighting coefficient of the neural network determined by executing the coefficient determination process. Thereafter, each time the coefficient update time arrives, the series of processes from steps S7 to S12 described above are repeatedly executed.

After that, as shown in FIG. 17, it is assumed that the cleaning worker of the accommodation facility H performs an operation on the terminal device 2 to update the cleaning worker's daily cleaning schedule. In this case, the terminal device 2 receives operation information indicating the details of the operation performed by the cleaning worker to update the cleaning schedule (step S13). Next, schedule request information requesting the cleaning load management device 1 to transmit schedule information is transmitted from the terminal device 2 to the cleaning load management device 1 (step S14). On the other hand, when the cleaning load management device 1 receives the schedule request information, the above-mentioned operation history request information and environmental performance request information are transmitted from the cleaning load management device 1 to the operating device 4 (step S15). On the other hand, when the operating device 4 receives the operation history request information and the environmental performance request information, the operating history information and the environmental performance information for the period covered by the cleaning load prediction are transmitted from the operating device 4 to the cleaning load management device 1. (step S16). Subsequently, the above-mentioned guest room information request information is transmitted from the cleaning load management device 1 to the customer server 10 (step S17). On the other hand, upon receiving the guest room information request information from the cleaning load management device 1, the customer server 10 specifies the guest room information corresponding to the received guest room information request information (step S18). Subsequently, the specified guest room information is transmitted from the customer server 10 to the cleaning load management device 1 (step S19). On the other hand, the cleaning load management device 1 uses a neural network to calculate a predicted cleaning load based on the received operation history information, environmental performance information, and guest room information (step S20). The cleaning load management device 1 causes the predicted load storage unit 135 to store predicted load information indicating the calculated predicted cleaning load.

Thereafter, the cleaning load management device 1 generates schedule information indicating the cleaning schedule of the cleaning worker based on the predicted load information stored in the predicted load storage section 135 and the worker information stored in the worker information storage section 137. is generated (step S21). Here, the cleaning load management device 1 stores the generated schedule information in the schedule storage unit 136. Next, the generated schedule information is transmitted from the cleaning load management device 1 to the terminal device 2 (step S22). On the other hand, upon receiving the schedule information, the terminal device 2 causes the display unit 204 to display an operation screen image for notifying the cleaning worker of the cleaning schedule indicated by the received schedule information (step S23).

Here, the terminal device 2 causes the display unit 204 to display operation screen images GA2 and GA3 for notifying the cleaning worker of the cleaning schedule as shown in FIG. 18(A) or (B), for example. As shown in FIG. 18(A), the operation screen image GA2 notifying the cleaning schedule of month and day displayed on the display unit 204 of the terminal device 2 owned by the cleaning worker is displayed between 9:00 and 9:30. It is set to clean guest room (room) XXXX from 9:30 to 9:45, to clean guest room (room) YYYY from 9:30 to 9:45, and to clean guest room (room) ZZZZ from 9:45 to 10:30. Here, the cleaning time for each room corresponds to the cleaning time required to return each room to its original state, and reflects the cleaning load determined by the neural network calculation unit 114. In this way, the display unit 204 of the terminal device 2 displays the cleaning schedule in which the cleaning time for each room is set to the optimal time according to the degree of dirt, improving the work efficiency of the cleaning worker. This reduces the waste of cleaning work. Further, as shown in FIG. 18(B), the operation screen image GA3 notifying the cleaning schedule for □ month □ day displayed on the display unit 204 of the terminal device 2 owned by the cleaning worker is displayed from 9:00 to 9:00 p.m. :45 to clean guest room (room) XXXX, from 9:450 to 10:15 to clean guest room (room) YYYY, and from 10:15 to 10:30 to clean guest room (room) ZZZZ. There is. The cleaning time for each room shown in FIG. 18(B) is different from the cleaning time for each room shown in FIG. 18(A). This reflects the fact that the degree of dirt in each room differs depending on how the air conditioner 5, lighting equipment 6, and bathroom ventilation fan 7 are used by the users staying in each room.

Next, the operation history information/environmental performance information management process executed by the operating device 4 according to the present embodiment will be described with reference to FIG. 19. This device status information transmission process is started, for example, when the operating device 4 is powered on. In addition, in the operating device 4, in parallel with the operation history information/environmental performance information management processing, the control information generation unit 412 generates control information according to the operation information received by the reception unit 411, and the control information transmission unit 414 generates control information. However, by transmitting the generated control information to the air conditioner 5, the lighting equipment 6, or the bathroom ventilation fan 7, a process of controlling the operation of the air conditioner 5, the lighting equipment 6, or the bathroom ventilation fan 7 is also executed.

First, the timer 419 determines whether a preset performance acquisition time has arrived (step S401). When the timer 419 determines that the performance acquisition time has not yet arrived (step S401: No), the process of step S405, which will be described later, is executed. On the other hand, when the timer 419 determines that the time to acquire the performance has arrived (step S401: Yes), the operation history acquisition unit 415 retrieves the control information from the control information storage unit 431, indicating the date and time when the control information was transmitted. Acquired along with date and time information. Then, the operation history acquisition unit 415 stores the combination of the acquired control information and date and time information in the operation history storage unit 432 as operation history information (step S402). Next, the environmental information acquisition unit 417 transmits the above-mentioned environmental request information to the air conditioner 5 (step S403). As a result, the environmental information acquisition unit 417 acquires environmental information from the air conditioner 5, and stores a combination of the acquired environmental information and date and time information indicating the date and time when the environmental information was acquired as environmental performance information. The information is stored in the unit 434 (step S404).

Subsequently, the operation history transmitting unit 416 and the environmental performance transmitting unit 418 determine whether or not the above-mentioned teacher information request information has been received from the cleaning load management device 1 (step S405). If the operation history transmitting unit 416 and the environmental performance transmitting unit 418 determine that teacher information request information has not been received (step S405: No), the process of step S408, which will be described later, is executed. On the other hand, assume that the operation history transmitting unit 416 and the environmental performance transmitting unit 418 determine that teacher information request information has been received (step S405: Yes). In this case, the operation history transmitting unit 416 transmits the operation history information within the above-mentioned teacher information target period stored in the operation history storage unit 432 to the cleaning load management device 1 (step S406), and the environmental performance transmitting unit 418 The environmental performance information within the above-mentioned teacher information target period stored in the environmental performance storage unit 434 is transmitted to the cleaning load management device 1 (step S407).

After that, the operation history transmitting unit 416 determines whether or not the above-mentioned operation history request information has been received from the cleaning load management device 1 (step S408). If the operation history transmitter 416 determines that the operation history request information has not been received (step S408: No), the process of step S410, which will be described later, is executed. On the other hand, if the operation history transmitting unit 416 determines that the operation history request information has been received (step S408: Yes), the operation history transmitting unit 416 transmits the operation history information within the above-mentioned cleaning load prediction target period stored in the operation history storage unit 432 to the cleaning load management. It is transmitted to device 1 (step S409). Next, the environmental performance transmitting unit 418 determines whether or not the above-mentioned environmental performance request information has been received from the cleaning load management device 1 (step S410). If the environmental performance transmitter 418 determines that the environmental performance request information has not been received (step S410: No), the process of step S401 is executed again. On the other hand, if the environmental performance transmitting unit 418 determines that the environmental performance request information has been received (step S410: Yes), the environmental performance information within the above-mentioned cleaning load prediction target period stored in the environmental performance storage unit 434 is used for cleaning load management. It is transmitted to device 1 (step S411). Subsequently, the process of step S401 is executed again.

Next, the terminal control process executed by the terminal device 2 according to the present embodiment will be described with reference to FIG. 20. This terminal control process is started, for example, when the terminal device 2 is powered on. First, the reception unit 211 determines whether or not operation information corresponding to the questionnaire input operation is received by the cleaning worker inputting the above-mentioned questionnaire regarding cleaning time and cleaning load on the input unit 205. (Step S201). At this time, the display control unit 212 acquires, for example, image information indicating the operation screen image GA1 shown in FIG. . Here, if the reception unit 211 determines that the operation information corresponding to the questionnaire input operation is not received (step S201: No), the process of step S203, which will be described later, is executed. On the other hand, if the receiving unit 211 determines that the operation information corresponding to the questionnaire input operation has been received (step S201: Yes), the load record acquisition unit 213 acquires the load record based on the operation information corresponding to the received questionnaire input operation. Information is generated and stored in the load record storage unit 232 (step S202).

Next, when the cleaning worker performs an operation on the input unit 205 to transmit the questionnaire response to the cleaning load management device 1, the receiving unit 211 determines whether or not operation information corresponding to the questionnaire sending operation has been received. (Step S203). Here, if the receiving unit 211 determines that the operation information corresponding to the questionnaire sending operation is not received (step S203: No), the process of step S205, which will be described later, is executed. On the other hand, if the receiving unit 211 determines that the operation information corresponding to the questionnaire sending operation has been received (step S203: Yes), the load record transmitting unit 214 transfers the load record information stored in the load record storage unit 232 to the cleaning load management device. 1 (step S204).

Subsequently, when the cleaning worker performs an operation on the input unit 205 to update the cleaning schedule stored in the schedule storage unit 231, the reception unit 211 determines whether or not operation information corresponding to the schedule update operation has been received. (Step S205). Here, if the reception unit 211 determines that the operation information corresponding to the schedule update operation is not received (step S205: No), the process of step S201 is executed again. On the other hand, assume that the reception unit 211 determines that operation information corresponding to a schedule update operation has been received (step S205: Yes). In this case, the schedule acquisition unit 215 acquires schedule information from the cleaning load management device 1 (step S206) by transmitting schedule request information requesting the cleaning load management device 1 to transmit schedule information (step S206). S207). At this time, the schedule acquisition unit 215 causes the schedule storage unit 231 to store the acquired schedule information. Thereafter, the display control unit 212 displays an operation screen image for notifying the cleaning worker of the cleaning schedule, for example, as shown in FIGS. 18(A) and 18(B), based on the schedule information stored in the schedule storage unit 231. The information is displayed on the display unit 204 (step S208). Then, the process of step S201 is executed again.

Next, the cleaning load management process executed by the cleaning load management device 1 according to the present embodiment will be described with reference to FIGS. 21 and 22. This cleaning load management process is started, for example, when the cleaning load management device 1 is powered on. First, the cleaning load acquisition unit 112 determines whether load performance information has been received from the terminal device 2 (step S101). If the cleaning load acquisition unit 112 determines that the load performance information has not been received (step S101: No), the process of step S103, which will be described later, is executed. On the other hand, when the cleaning load acquisition unit 112 determines that the load performance information has been received (step S101: Yes), the cleaning load acquisition unit 112 stores the received load performance information in the cleaning load storage unit 132 (step S102).

Next, the timer 120 determines whether a preset coefficient update time has arrived (step S103). If the timer 120 determines that the coefficient update time has not yet arrived (step S103: No), the process of step S109, which will be described later, is executed. On the other hand, when the timer unit 120 determines that the coefficient update time has arrived (step S103: Yes), the operation history acquisition unit 111 and the environmental performance acquisition unit 119 each send the above-mentioned teacher information request information to the operating device 4. Transmit (step S104). As a result, the operation history acquisition unit 111 acquires the operation history information within the teacher information target period from the operation device 4, and the environmental performance acquisition unit 119 acquires the environmental performance information within the teacher information target period from the operation device 4. Acquire (step S105). Here, the operation history acquisition unit 111 stores the acquired operation history information in the operation history storage unit 131, and the environmental performance acquisition unit 119 stores the acquired environmental performance information in the environmental performance storage unit 138. Subsequently, the guest room information acquisition unit 113 transmits teacher information request information to the customer server 10 (step S106), thereby acquiring guest room information within the teacher information target period from the customer server 10 (step S107). Here, the guest room information acquisition section 113 causes the guest room information storage section 133 to store the acquired guest room information. After that, coefficient determination processing is executed (step S108).

Here, details of the coefficient determination process will be explained in detail with reference to FIG. 22. First, the neural network calculation unit 114 acquires operation history information from the operation history storage unit 131, environmental performance information from the environmental performance storage unit 138, and guest room information from the guest room information storage unit. Further, the neural network calculation unit 114 acquires load performance information from the cleaning load storage unit 132 (step S301). Here, the operation history information, environmental performance information, guest room information, and load performance information are all information for the aforementioned teacher information target period. Next, the coefficient setting unit 116 acquires initial weighting coefficient information indicating the initial weighting coefficient or weighting coefficient information indicating the most recently used weighting coefficient from the neural network storage unit 134, and the weighting coefficient indicated by the acquired weighting coefficient information. is set as the weighting coefficient of the neural network (step S302). Next, the neural network calculation unit 114 calculates the predicted cleaning load from the acquired operation history information, environmental performance information, and guest room information using the neural network for which the weighting coefficient has been set by the coefficient setting unit 116 (step S303 ). After that, the coefficient determining unit 115 calculates the error between the calculated predicted cleaning load and the cleaning load indicated by the load performance information (step S304). Next, the coefficient determination unit 115 determines a new weighting coefficient for the neural network by backpropagation based on the calculated error (step S305). Then, the coefficient determination unit 115 stores the determined weighting coefficients in the neural network storage unit 134 (step S306).

Returning to FIG. 21, next, the schedule generation unit 117 determines whether schedule update request information has been received from the terminal device 2 (step S109). If the schedule generation unit 117 determines that schedule update request information has not been received from the terminal device 2 (step S109: No), the process of step S101 is executed again. On the other hand, assume that the schedule generation unit 117 determines that schedule update request information has been received from the terminal device 2 (step S109: Yes). In this case, the operation history acquisition unit 111 transmits the above-mentioned operation history request information to the operating device 4, and the environmental performance acquisition unit 119 transmits the above-mentioned environmental performance request information to the operating device 4 (step S110). As a result, the operation history acquisition unit 111 acquires the operation history information within the above-mentioned cleaning load prediction period from the operation device 4, and the environmental performance acquisition unit 119 acquires the environmental performance information within the cleaning load prediction period from the operation device 4. Information is acquired (step S111). Here, the operation history acquisition unit 111 stores the acquired operation history information in the operation history storage unit 131, and the environmental performance acquisition unit 119 stores the acquired environmental performance information in the environmental performance storage unit 138. Subsequently, the guest room information acquisition unit 113 transmits guest room information request information to the customer server 10 (step S112), thereby acquiring guest room information within the cleaning load prediction period from the customer server 10 (step S113). Here, the guest room information acquisition section 113 causes the guest room information storage section 133 to store the acquired guest room information.

Thereafter, the neural network calculation unit 114 calculates the predicted cleaning load from the acquired operation history information, environmental performance information, and guest room information using the neural network whose weighting coefficients have been set by the coefficient setting unit 116 (step S114). . Here, the neural network calculation unit 114 causes the predicted load storage unit 135 to store cleaning load information indicating the calculated predicted cleaning load.

Next, the schedule generation unit 117 selects a cleaning worker based on the cleaning load calculated by the neural network calculation unit 114 stored in the predicted load storage unit 135 and the worker information stored in the worker information storage unit 137. Schedule information indicating a cleaning schedule is generated (step S115). Here, the schedule generation unit 117 causes the schedule storage unit 136 to store the generated schedule information. Subsequently, the schedule transmitting unit 118 transmits the schedule information stored in the schedule storage unit 136 to the terminal device 2 (step S116). After that, the process of step S101 is executed again.

As explained above, according to the cleaning load management device 1 according to the present embodiment, the neural network calculation unit 114 uses the neural network whose weighting coefficient has been determined by the coefficient determination unit 115 to calculate the cleaning load prediction target period. The cleaning load when cleaning the guest room RA (RB) is determined from the operation history information, environmental performance information, and guest room information. As a result, it is possible to estimate the cleaning load according to the usage status of the guest room RA (RB) by the user and the degree of dirt in the room after the user stays in the guest room RA (RB). The optimal cleaning work time required to clean and restore the original state can be calculated and reflected in the cleaning worker's cleaning schedule. Therefore, wasted time in the cleaning schedule can be reduced, and the efficiency of cleaning work can be improved.

In addition, the guest room information acquisition unit 113 according to the present embodiment also determines the size of the guest rooms RA and RB, the type of user who stayed in the guest rooms RA and RB, that is, whether the user is an adult or a child, and the size of the guest rooms RA and RB. Acquire room information including the number of users who stayed. As a result, the neural network calculation unit 114 calculates the cleaning load based on the size of the guest rooms RA and RB, the types of users who stayed in the guest rooms RA and RB, and the number of users who stayed in the guest rooms RA and RB. be able to. Therefore, the accuracy of the calculated cleaning load increases, making it possible to create an appropriate cleaning schedule accordingly.

Further, the schedule generation unit 117 according to the present embodiment generates schedule information based on usage status information indicating whether the user is staying in guest rooms RA and RB. As a result, the cleaning schedule is created taking into account the time period when the guest rooms RA and RB cannot be cleaned because the user is staying, so the wasted time period in the cleaning schedule is reduced.

(Embodiment 2)
The cleaning load management system according to this embodiment includes a cleaning load management device installed in each of a plurality of accommodation facilities, and a cloud server that can communicate with the plurality of cleaning load management devices via a wide area network. Then, in the cloud server, the neural network coefficients of the neural network used in each cleaning load management device for determining the cleaning load when cleaning a room are determined.

For example, as shown in FIG. 23, the cleaning load management system according to the present embodiment includes a cleaning load management device 2001 installed in each of a plurality of accommodation facilities H1, H2, H3, . . . and a terminal device 2. An operating device 4 is provided. The cleaning load management system also includes a cloud server 2100 and a customer server 10 that can communicate with the cleaning load management device 1 via the wide area network NW1. Note that in FIG. 23, the same components as in Embodiment 1 are given the same reference numerals as in FIG.

The cleaning load management device 2001 has the same hardware configuration as the cleaning load management device 1 described in Embodiment 1, and as shown in FIG. , a local communication interface 104, and a bus 109 that connects these to each other. Then, by the CPU 101 reading out the program stored in the auxiliary storage unit 103 to the main storage unit 102 and executing it, the operation history acquisition unit 111, the guest room information acquisition unit 113, the environmental performance acquisition unit 119 , a neural network calculation section 114, a teacher information transmission section 2122, a coefficient acquisition section 2121, a coefficient setting section 116, a schedule generation section 117, a schedule transmission section 118, and a time measurement section 2120. Note that in FIG. 24, the same components as in Embodiment 1 are given the same reference numerals as in FIG. The auxiliary storage unit 103 shown in FIG. 6 also includes an operation history storage unit 131, a guest room information storage unit 133, an environmental performance storage unit 138, a neural network storage unit 134, and a predicted load storage unit, as shown in FIG. 135, a schedule storage section 136, and a worker information storage section 137.

The teacher information transmitting unit 2122 acquires operation history information within the teacher information target period from the operation history storage unit 131 and acquires environmental performance information within the teacher information target period from the environmental performance storage unit 138. Then, the teacher information transmitting unit 2122 transmits the acquired operation history information and environmental performance information to the cloud server 2100 as teacher information for determining the weighting coefficient of the neural network for determining the cleaning load.

The coefficient acquisition unit 2121 acquires coefficient information indicating the weighting coefficient of the neural network determined by the cloud server 2100 from the cloud server 2100. Then, the coefficient acquisition unit 2121 causes the neural network storage unit 134 to store weighting coefficient information included in the acquired coefficient information.

The timer unit 2120 determines whether the coefficient update time for updating the weighting coefficients of the neural network stored in the neural network storage unit 134 has arrived, based on time information output from a software timer (not shown). . When the coefficient update time arrives, the clock unit 2120 outputs coefficient update time arrival notification information to the operation history acquisition unit 111, the environmental performance acquisition unit 119, the guest room information acquisition unit 113, and the teacher information transmission unit 2122.

For example, as shown in FIG. 25, the cloud server 2100 includes a CPU 1001, a main storage unit 1002, an auxiliary storage unit 1003, a wide area communication interface 1004, and a bus 1009 that connects these to each other. CPU 1001 is, for example, a multi-core processor. The main storage unit 1002 is used as a work area for the CPU 101, and the auxiliary storage unit 1003 stores programs for the CPU 1001 to execute various processes. The wide area communication interface 1004 is connected to the wide area network NW1, and is capable of communicating with the cleaning load management devices 2001 installed in each of the plurality of accommodation facilities H1, H2, H3, . . . via the wide area network NW1. .

By reading the program stored in the auxiliary storage unit 1003 into the main storage unit 1002 and executing it, the CPU 1001 acquires the teacher information acquisition unit 2111, the cleaning load acquisition unit 2118, the guest room information acquisition unit 2112, and the neural network, as shown in FIG. It functions as a network calculation section 2113, a coefficient determination section 2115, a coefficient setting section 2114, a coefficient information generation section 2116, and a coefficient information transmission section 2117. Further, the auxiliary storage unit 1003 shown in FIG. 25 includes a teacher information storage unit 2131, a cleaning load storage unit 2134, a guest room information storage unit 2132, and a neural network storage unit 2133, as shown in FIG. The teacher information storage unit 2131 stores information indicating the operating status of the air conditioner 5, lighting equipment 6, and bathroom ventilation fan 7 installed in each of the accommodation facilities H1, H2, H3, . . . H1, H2, H3, . . . are stored in association with accommodation facility identification information and room identification information.

In addition, the teacher information storage unit 2131 stores environmental performance information indicating past performance values of environmental parameters such as temperature and humidity in each room (for example, guest rooms RA and RB) of the accommodation facilities H1, H2, H3, etc. , date and time information, room identification information, and accommodation facility identification information are stored in association with each other. The cleaning load storage unit 2134 stores load performance information including work time index information representing an index of the length of cleaning work time and workload index information representing an index of the load of cleaning work, and room identification information and cleaning work. It is stored in association with work day information indicating the day on which the work was carried out and accommodation facility identification information.

The guest room information storage unit 2132 stores accommodation facility identification information, room type information indicating the type of each of a plurality of rooms in the accommodation facility H, room area information indicating the size of the room, and the type of user staying in the room. and the number of guests staying in the room, and room information including the number of guests staying in the room.

The neural network storage unit 2133 stores information regarding the same neural network as the neural network used in the cleaning load management device 2001 installed in each of the accommodation facilities H1, H2, H3, . . . . Specifically, the neural network storage unit 2133 stores weighting coefficient information indicating weighting coefficients of the neural network.

The teacher information acquisition unit 2111 acquires teacher information including operation history information and environmental performance information from the cleaning load management device 2001 installed in each of the accommodation facilities H1, H2, H3, . . . . Here, the operation history information is information indicating the operation history of each of the air conditioners 5, lighting fixtures 6, and bathroom ventilation fans 7 installed in each guest room RA, RB of the accommodation facilities H1, H2, H3, . . . . Further, the environmental performance information is information indicating the environmental performance of each guest room RA, RB of the accommodation facilities H1, H2, H3, . . . . The guest room information acquisition unit 2112 acquires guest room information from the customer server 10 by transmitting teacher information request information to the customer server 10. The guest room information acquisition section 2112 causes the guest room information storage section 2132 to store the acquired guest room information. The cleaning load acquisition unit 2118 acquires load performance information indicating the cleaning load, which is the workload required for cleaning guest rooms RA and RB of the accommodation facility H, from the terminal device 2 . The cleaning load acquisition unit 2118 causes the cleaning load storage unit 2134 to store the acquired load performance information.

The coefficient determination unit 2115 determines the weighting coefficient of the neural network for determining the cleaning load described above, based on the operation history information, environmental performance information, guest room information, and load performance information. First, the coefficient setting unit 2114 acquires initial weighting coefficient information from the neural network storage unit 2133, and sets the initial weighting coefficient indicated by the acquired initial weighting coefficient information as the weighting coefficient of the neural network. Next, the coefficient determination unit 2115 obtains the cleaning load calculated by the neural network calculation unit 2113 using the neural network. Here, the neural network calculation unit 2113 calculates the cleaning load using a neural network based on the operation history information, environmental performance information, and guest room information, which are teacher information. Next, the coefficient determination unit 2115 acquires load performance information, which is teacher information stored in the cleaning load storage unit 2134, and calculates the difference between the cleaning load indicated by the acquired load performance information and the cleaning load calculated using the neural network. Calculate the error. Then, the coefficient determination unit 2115 determines the weighting coefficient of the neural network using the error backpropagation method based on the calculated error. Further, the coefficient determining unit 2115 causes the neural network storage unit 2133 to store weighting coefficient information indicating the determined weighting coefficient.

The coefficient information generation unit 2116 acquires weighting coefficient information indicating the weighting coefficient determined by the coefficient determining unit 2115 from the neural network storage unit 2133, and generates coefficient information including the acquired weighting coefficient information. The coefficient information transmitter 2117 transmits the coefficient information generated by the coefficient information generator 2116 to the cleaning load management device 2001.

Next, the operation of the cleaning load management system according to this embodiment will be described with reference to FIGS. 27 to 29. First, as shown in FIG. 27, when the preset performance information acquisition time arrives, the operating device 4 executes the processes of steps S21 and S22. The processing in steps S21 and S22 is similar to the processing in steps S1 and S2 described in the first embodiment. Thereafter, the operating device 4 repeatedly executes the processes of steps S21 and S22 every time the performance information acquisition time arrives. In addition, after cleaning the guest room RA (RB), the cleaning worker of the accommodation facility H cleans the guest room RA (RB), and with the operation screen image for inputting a questionnaire regarding the cleaning load displayed on the display unit 204 of the terminal device 2, the cleaning worker cleans the guest room RA (RB). Suppose that the user performs an operation of inputting information into the questionnaire via the input unit 205 of No. 2. In this case, the terminal device 2 receives operation information indicating the contents of the input operation to the questionnaire (step S23). Next, the terminal device 2 generates load record information indicating the cleaning load record corresponding to the operation information indicating the received operation details, and stores it in the load record storage unit 232 (step S24). Subsequently, when the terminal device 2 receives an operation for transmitting the questionnaire results after the cleaning worker inputs the questionnaire, the load record information stored in the load record storage unit 232 is stored in the terminal device 2. from there to the cloud server 2100 (step S25). On the other hand, upon receiving the load performance information, the cloud server 2100 stores the received load performance information in the teacher information storage unit 2131 (step S26). Thereafter, each time the cleaning worker performs an input operation to the questionnaire, the series of processes from steps S23 to S26 described above are repeatedly executed.

After that, it is assumed that the time has come to update the preset weighting coefficients of the neural network used by the cleaning load management device 1. In this case, teacher information request information that requests the operating device 4 to transmit operation history information and environmental performance information as teacher information is transmitted from the cleaning load management device 2001 to the operating device 4 (step S27). On the other hand, when the operating device 4 receives the teacher information request information, operation history information and environmental performance information are transmitted from the operating device 4 to the cleaning load management device 2001 (step S28). Here, the operation history information and environmental performance information are the same as the operation history information and environmental performance information described in the first embodiment.

Next, as shown in FIG. 28, teacher information including operation history information and environmental performance information acquired by the cleaning load management device 2001 from the operating device 4 is transmitted from the cleaning load management device 2001 to the cloud server 2100 (step S29). Here, upon receiving the teacher information, the cloud server 2100 causes the teacher information storage unit 2131 to store the operation history information and environmental performance information included in the received teacher information. Subsequently, teacher information request information requesting the customer server 10 to transmit guest room information as teacher information is transmitted from the cloud server 2100 to the customer server 10 (step S30). On the other hand, upon receiving the teacher information request information from the cloud server 2100, the customer server 10 identifies guest room information corresponding to the received teacher information request information (step S31). Subsequently, the specified guest room information is transmitted from the customer server 10 to the cloud server 2100 (step S32). On the other hand, the cloud server 2100 executes a coefficient determination process to determine the weighting coefficient of the neural network based on the received guest room information, operation history information, environmental performance information, and load performance information stored in the teacher information storage unit 2131. (Step S33). The cloud server 2100 causes the neural network storage unit 2133 to store weighting coefficient information indicating the weighting coefficient of the neural network determined by executing the coefficient determination process. Thereafter, coefficient information including weighting coefficient information stored in the neural network storage unit 2133 is transmitted from the cloud server 2100 to the cleaning load management device 2001 (step S34). On the other hand, upon receiving the coefficient information, the cleaning load management device 2001 updates the weighting coefficient information by storing the weighting coefficient included in the received coefficient information in the neural network storage unit 134 (step S35). Thereafter, each time the coefficient update time arrives, the series of processes from steps S27 to S35 described above are repeatedly executed.

Next, assume that the cleaning worker of the accommodation facility H performs an operation on the terminal device 2 to update the cleaning worker's daily cleaning schedule. In this case, the terminal device 2 receives operation information indicating the details of the operation performed by the cleaning worker to update the cleaning schedule (step S36). Subsequently, schedule request information is transmitted from the terminal device 2 to the cleaning load management device 2001 (step S37). On the other hand, when the cleaning load management device 2001 receives the schedule request information, the above-mentioned operation history request information and environmental performance request information are transmitted from the cleaning load management device 2001 to the operating device 4 (step S38). On the other hand, when the operating device 4 receives the operation history request information and the environmental performance request information, the operating history information and the environmental performance information for the cleaning load prediction period are transmitted from the operating device 4 to the cleaning load management device 1 in response to the operation history request information and the environmental performance request information. (step S39). Thereafter, as shown in FIG. 29, the above-mentioned guest room information request information is transmitted from the cleaning load management device 2001 to the customer server 10 (step S40).

On the other hand, upon receiving the guest room information request information from the cleaning load management device 2001, the customer server 10 identifies guest room information corresponding to the received guest room information request information (step S41). Next, the identified guest room information is transmitted from the customer server 10 to the cleaning load management device 2001 (step S42). On the other hand, the cleaning load management device 2001 uses a neural network to calculate a predicted cleaning load based on the received operation history information, environmental performance information, and guest room information (step S43). Thereafter, the cleaning load management device 2001 generates schedule information indicating the cleaning schedule of the cleaning worker based on the predicted load information stored in the predicted load storage unit 135 and the worker information stored in the worker information storage unit 137. is generated (step S44). Next, the generated schedule information is transmitted from the cleaning load management device 2001 to the terminal device 2 (step S45). On the other hand, upon receiving the schedule information, the terminal device 2 causes the display unit 204 to display an operation screen image for notifying the cleaning worker of the cleaning schedule indicated by the received schedule information (step S46).

Next, the cleaning load management process executed by the cleaning load management device 2001 according to this embodiment will be described with reference to FIG. 30. First, the timer 2120 determines whether a preset coefficient update time has arrived (step S2101). If the timer 2120 determines that the coefficient update time has not yet arrived (step S2101: No), the process of step S2106, which will be described later, is executed. On the other hand, when the timer unit 2120 determines that the coefficient update time has arrived (step S2101: Yes), the operation history acquisition unit 111 and the environmental performance acquisition unit 119 each send the above-mentioned teacher information request information to the operating device 4. Send (step S2102). As a result, the operation history acquisition unit 111 acquires the operation history information within the teacher information target period from the operating device 4, and the environmental performance acquisition unit 119 acquires the environmental performance information within the teacher information target period from the operation device 4. Acquire (step S2103). Next, the teacher information transmitter 2122 transmits teacher information including the operation history information and environmental performance information acquired by the operation history acquisition unit 111 and the environmental performance acquisition unit 119 to the cloud server 2100 (step S2104). Subsequently, the coefficient acquisition unit 2121 acquires coefficient information from the cloud server 2100, and stores the weighting coefficient information included in the acquired coefficient information in the neural network storage unit 134 (step S2105).

Thereafter, the schedule generation unit 117 determines whether schedule update request information has been received from the terminal device 2 (step S2106). If the schedule generation unit 117 determines that schedule update request information has not been received from the terminal device 2 (step S2106: No), the process of step S2101 is executed again. On the other hand, assume that the schedule generation unit 117 determines that schedule update request information has been received from the terminal device 2 (step S2106: Yes). In this case, the operation history acquisition unit 111 transmits the above-mentioned operation history request information to the operating device 4, and the environmental performance acquisition unit 119 transmits the above-mentioned environmental performance request information to the operating device 4 (step S2107). As a result, the operation history acquisition unit 111 acquires the operation history information within the above-mentioned cleaning load prediction period from the operation device 4, and the environmental performance acquisition unit 119 acquires the environmental performance information within the cleaning load prediction period from the operation device 4. Information is acquired (step S2108). Subsequently, the guest room information acquisition unit 113 transmits guest room information request information to the customer server 10 (step S2109), thereby acquiring guest room information within the cleaning load prediction period from the customer server 10 (step S2110).

Next, the neural network calculation unit 114 calculates a predicted cleaning load from the acquired operation history information, environmental performance information, and guest room information using the neural network for which the weighting coefficient has been set by the coefficient setting unit 116 (step S2111 ). Next, the schedule generation unit 117 generates schedule information indicating the cleaning schedule of the cleaning worker based on the cleaning load calculated by the neural network calculation unit 114 and the worker information stored in the worker information storage unit 137. is generated (step S2112). Thereafter, the schedule transmitter 118 transmits the generated schedule information to the terminal device 2 (step S2113). After that, the process of step S2101 is executed again.

Next, the coefficient information generation process executed by cloud server 2100 according to this embodiment will be described with reference to FIG. 31. First, the cleaning load acquisition unit 2118 determines whether load performance information has been received from the terminal device 2 (step S21001). If the cleaning load acquisition unit 112 determines that load performance information has not been received (step S21001: No), the process of step S21003, which will be described later, is executed. On the other hand, if the cleaning load acquisition unit 112 determines that the load performance information has been received (step S21001: Yes), the cleaning load acquisition unit 112 stores the received load performance information in the cleaning load storage unit 132 (step S21002).

Next, the teacher information acquisition unit 2111 determines whether teacher information has been received from the cleaning load management device 2001 (step S21003). If the teacher information acquisition unit 2111 determines that teacher information has not been received (step S21003: No), the process of step S21001 is executed again. On the other hand, when the teacher information acquisition unit 2111 determines that the teacher information has been received (step S21003: Yes), the teacher information acquisition unit 2111 stores the operation history information and environmental performance information within the teacher information target period included in the received teacher information in the teacher information storage unit. 2131. Subsequently, the guest room information acquisition unit 2112 transmits teacher information request information to the customer server 10 (step S21004), and acquires guest room information within the teacher information target period from the customer server 10 (step S21005). Here, the guest room information acquisition section 2112 causes the guest room information storage section 2132 to store the acquired guest room information. After that, coefficient determination processing is executed (step S21006). This coefficient determination process is similar to the coefficient determination process described using FIG. 22 in the first embodiment. Here, the coefficient determining unit 2115 causes the neural network storage unit 2133 to store weighting coefficient information indicating the determined weighting coefficient.

Next, the coefficient information generation unit 2116 generates coefficient information including the weighting coefficient information stored in the neural network storage unit 2133 (step S21007). Subsequently, the coefficient information transmitting unit 2117 transmits the generated coefficient information to the cleaning load management device 2001 (step S21008). After that, the process of step S21001 is executed again.

As described above, according to the control system according to the present embodiment, the weighting coefficient of the neural network used in the cleaning load management device 2001 is determined in the cloud server 2100. This has the advantage that the processing load on the cleaning load management device 2001 is reduced.

Although each embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiments. For example, as shown in FIG. 32, when the operating device 4 displays a language selection button image BU1 for selecting the language of the message (for example, messages M31, M32) on the display unit 404 together with a message, the input unit 405 The operation history information may include language information indicating the language selected by touching the portion corresponding to the language selection button image BU1. In the example shown in FIG. 32, a language selection button image BU1 is arranged at the lower right of the display section 404. Here, when the language of the message displayed on the display unit 404 is Japanese, the language selection button image BU1 is displayed as "Eng". Then, when a portion of the input section 405 corresponding to the language selection button image BU1 displayed as "Eng" is touched, the language of the message displayed on the display section 404 is switched to English. At this time, the language selection button image BU1 is displayed as "Chi". Then, when a portion of the input section 405 corresponding to the language selection button image BU1 displayed as "Chi" is touched, the language of the message displayed on the display section 404 is switched to Chinese. In this way, by touching the part corresponding to the language selection button image BU1 on the input section 405, the language of the message displayed on the display section 404 can be changed from "Japanese" to "English" to "Chinese". You can switch in the order of →・・・・→“Japanese”. Then, the operating device 4 may transmit information indicating a history of operations of touching a portion of the input unit 405 corresponding to the language selection button image BU1 to the cleaning load management device 1 as operation history information.

Here, the influence of the language selection of the users of guest rooms RA and RB on the amount of cleaning work will be explained. The usage, usage time, and frequency of use of the air conditioners 5, lighting fixtures 6, and bathroom ventilation fans 7 installed in guest rooms RA and RB tend to differ depending on the lifestyle or national characteristics of the users. For example, Westerners tend not to use bathtubs, but only showers. In this case, the amount of time the user stays in the bathroom becomes shorter, so the degree of dirt in the bathroom becomes smaller, so the cleaning load when cleaning the inside of the bathroom becomes smaller. On the other hand, Japanese people tend to use not only showers but also bathtubs. In this case, the user stays in the bathroom for a longer time, so the degree of dirt in the bathroom tends to be higher than that of Westerners. Therefore, the cleaning load when cleaning the bathroom increases. Then, by adding information indicating the history of the operation of touching the part corresponding to the language selection button image BU1 on the input unit 405, which reflects the language selected on the operating device 4, to the operation history information, Since the user's lifestyle or national character can be reflected in the predicted cleaning load, the accuracy of the predicted cleaning load can be increased accordingly.

In each of the embodiments, the neural network calculation unit 114 calculates the operating history information from the operation history information in the period from the beginning of the cleaning load prediction target period to a point in time past the end of the cleaning load prediction target period by a preset margin period. The cleaning load when cleaning RA and RB may be determined. For example, if the cleaning load prediction target period is set from the time the user checks in to the time the user checks out, the neural network calculation unit 114 performs checkout from the beginning of the cleaning load prediction target period. The cleaning load may be calculated based on operation history information and environmental performance information up to a point 10 minutes earlier than the time when the cleaning load is set.

According to this configuration, the cleaning schedule can be created before the user actually checks out of the accommodation facility H, so the work efficiency of the cleaning worker can be increased accordingly.

In each embodiment, an example has been described in which the environmental performance acquisition unit 119 acquires environmental information indicating environmental parameters such as temperature and humidity of a place where the air conditioner 5 is installed. However, the type of environmental information is not limited to this, and indicates environmental parameters such as the illuminance of the place where the lighting fixture 6 is installed, and the temperature, humidity, and illuminance of the bathroom where the bathroom ventilation fan 7 is installed. It may also be something that acquires environmental information. In this case, for example, if the bathroom ventilation fan 7 has a temperature sensor and a humidity sensor, and the environmental performance acquisition unit 119 acquires environmental information indicating the temperature and humidity of the bathroom detected by the temperature sensor and humidity sensor. good. Further, the lighting fixture 6 may include an illuminance sensor, and the environmental performance acquisition unit 119 may acquire environmental information indicating the illuminance of the bedroom or bathroom detected by the illuminance sensor.

For example, if the air conditioner 5 is turned on and off frequently in a state where there is little change in the indoor temperature detected by the temperature sensor and humidity sensor, the amount of activity of the user tends to be large, and the user The weighting coefficient of the neural network is determined so that the cleaning load of guest rooms RA and RB to be cleaned becomes large. On the other hand, when the indoor temperature detected by the temperature sensor and the humidity sensor is changing within a temperature range where the user can feel comfortable, for example between 18 degrees Celsius and 25 degrees Celsius, the user turns off the air conditioner 5. If the user does not turn on the lighting equipment 6 when the indoor illuminance detected by the illuminance sensor is changing between 50 lux and 150 lux, for example, the user does not turn on the lighting equipment 6. The weight coefficients of the neural network are determined so that the cleaning load of the guest rooms RA and RB occupied by the user tends to be high, and the cleaning load of the guest rooms RA and RB occupied by the user tends to be high.

Alternatively, in each embodiment, the neural network calculation unit 114 uses a neural network to calculate the cleaning load when cleaning the guest room RA (RB) based only on the operation history information and guest room information during the cleaning load prediction period. It may also be something that requires.

In the second embodiment, the neural network calculation unit 114 calculates, in addition to the above-mentioned operation history information, environmental performance information, etc., regional information indicating areas where accommodation facilities H1, H2, H3, etc. are located, and accommodation facilities. The cleaning load may be calculated based on at least one of management entity identification information that identifies the management entity that manages H1, H2, H3, . . . . Here, the regional information indicates, for example, a region-specific regional identification number assigned to each region, and the management entity identification information indicates, for example, a unique regional identification number assigned to each of the management entities of accommodation facilities H1, H2, H3, etc. It may also indicate the management entity identification number of the company. In this case, the cleaning load management device 2001 may include at least one of a local information acquisition unit that acquires local information from the customer server 10 and a management entity identification information acquisition unit that acquires management entity identification information. good. Further, the cloud server 2100 may also include at least one of a regional information acquisition section that acquires regional information from the customer server 10 and a business entity identification information acquisition section that acquires business entity identification information. Then, the coefficient determination unit 2115 may determine the weighting coefficient of the neural network based on the regional information and management entity identification information acquired from the customer server 10.

In each embodiment, the schedule generation unit 117 calculates the cleaning load for the guest rooms RA and RB and the length of service and experience period of each cleaning worker who is working on the day when the cleaning work for the guest rooms RA and RB is to be performed. An example has been described in which the cleaning work time for each guest room RA and RB is calculated based on the length. However, the present invention is not limited to this, and for example, the schedule generation unit 117 may create a schedule for each guest room RA, RB based on the gender, age, and number of guest rooms of each cleaning worker who is working on the day the guest room RA, RB cleaning work is to be performed. The method may also be used to calculate the cleaning work time.

In the cleaning load management device 1, 2001 according to each embodiment, the guest room information acquisition unit 113, 2112 acquires the actual check-in time and check-out time of the users of guest rooms RA and RB, and obtains the guest room information. It may be stored in the storage unit 133 or 2132. Here, for example, if the time to operate the air conditioner 5, lighting equipment 6, or bathroom ventilation fan 7 in guest rooms RA and RB is short compared to the length of stay in guest rooms RA and RB, The weighting coefficient is determined so that the cleaning load for guest rooms RA and RB used by the user is small.

In each embodiment, the cleaning load management device 1, 2001 includes a seasonal information generation unit (not shown) that generates seasonal information indicating the season to which the date and time indicated by the date and time information belongs based on the date and time information. It's okay. Here, the seasonal information generation unit assigns seasonal information "10" indicating "spring" when the date and time indicated by the date and time information is included in the period from March to June, for example, and when the date and time indicated by the date and time information is included in July or June. If the month falls in August, seasonal information "20" indicating "summer" may be added. Further, the seasonal information generation unit assigns seasonal information "30" indicating "autumn" when the date and time indicated by the date and time information is included in the period from September to November, and when the date and time indicated by the date and time information is included in December or February. If it is included in the month, seasonal information "40" indicating "winter" may be added. In this case, the neural network calculation unit 114 may calculate the cleaning load based on the operation history information, environmental performance information, guest room information, and seasonal information.

Further, various functions of the cleaning load management device 1, 2001 and the cloud server 2100 according to the present invention may be realized by software, firmware, or a combination of software and firmware. In this case, the software or firmware is written as a program, and the program is stored on a computer-readable disk such as a flexible disk, CD-ROM (Compact Disc Read Only Memory), DVD (Digital Versatile Disc), and MO (Magneto-Optical Disc). A computer that can realize each of the above-mentioned functions may be configured by storing and distributing the program on a recording medium, reading the program into the computer, and installing the program. When each function is realized by sharing between an OS (Operating System) and an application, or by cooperation between the OS and an application, only the parts other than the OS may be stored in the recording medium.

Furthermore, it is also possible to superimpose each program on a carrier wave and distribute it via a network. For example, the program may be posted on a bulletin board (BBS, Bulletin Board System) on a network and distributed via the network. The above-described process may be executed by starting these programs and executing them under the control of the OS in the same way as other application programs.

INDUSTRIAL APPLICATION This invention is suitable for the system which creates the cleaning schedule of each room of an accommodation facility.

1,2001,3001 Cleaning load management device, 2 Terminal device, 4 Operating equipment, 5 Air conditioner, 6 Lighting equipment, 7 Bathroom ventilation fan, 8 Wireless communication device, 10 Customer server, 91 Router, 92 Data line termination device, 101 , 201, 401, 501, 1001 CPU, 102, 202, 402, 502, 1002 Main storage section, 103, 203, 403, 503, 1001 Auxiliary storage section, 104 Local communication interface, 105, 305 External communication interface, 109 , 209, 409, 509, 1009 Bus, 111, 415 Operation history acquisition unit, 112 Cleaning load acquisition unit, 113, 2112 Guest room information acquisition unit, 114, 2113 Neural network calculation unit, 115, 2115 Coefficient determination unit, 116, 2114 coefficient setting unit, 117 schedule generation unit, 118 schedule transmission unit, 119 environmental performance acquisition unit, 131, 432 operation history storage unit, 132 cleaning load storage unit, 133, 2132 guest room information storage unit, 134, 2133 neural network storage unit, 135 predicted load storage unit, 136, 231 schedule storage unit, 137 worker information storage unit, 138, 434 environmental performance storage unit, 204, 404 display unit, 205, 405 input unit, 206 wireless module, 211, 411 reception unit, 212, 413 Display control unit, 213 Load performance acquisition unit, 214 Load performance transmission unit, 215 Schedule acquisition unit, 232 Load performance storage unit, 233, 433 Image storage unit, 407, 507 Device communication interface, 412 Control information generation unit, 414 control information transmitter, 416 operation history transmitter, 417 environmental information acquirer, 418 environmental performance transmitter, 431 control information storage unit, 506 local communication interface, 511 operation controller, 512 environmental information generator, 513 relay unit, 2111 Teacher information acquisition unit, 2116 Coefficient information generation unit, 2117 Coefficient information transmission unit, 2131 Teacher information storage unit, BA1, BA2 Touch area, GA1, GA2, GA3, GA4 Operation screen image, H, H1, H2, H3 Accommodation facility , M1, MA1, MA2 area, MQ1, MQ2 message, NW1 wide area network, NW2 local network, RA, RB guest room

Claims (10)

an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in the room;
a cleaning load acquisition unit that acquires load performance information indicating a cleaning load that is a workload required for cleaning the room;
a coefficient determination unit that determines a neural network coefficient of a neural network for determining a cleaning load when cleaning the room after a user stays in the room, based on the operation history information and the load performance information; ,
Using the neural network for which the neural network coefficients have been determined by the coefficient determining unit, the neural network of the room is determined based on operation history information during a cleaning load prediction target period that includes at least a portion of the period during which the user stayed in the room. a neural network calculation unit that calculates a cleaning load when cleaning;
Cleaning load management device. further comprising an environmental performance acquisition unit that acquires environmental performance information indicating the environmental performance of the room,
The coefficient determination unit determines the neural network coefficient based on the environmental performance information as well as the operation history information and the load performance information,
The neural network calculation unit uses the neural network whose neural network coefficients have been determined by the coefficient determination unit to perform cleaning for the room based on the operation history information and the environmental performance information during the cleaning load prediction period. Find the cleaning load when cleaning,
The cleaning load management device according to claim 1. further comprising a guest room information acquisition unit that acquires guest room information including at least one of the size of the room, the type of users who stayed in the room, and the number of users who stayed in the room,
The coefficient determination unit determines the neural network coefficient based on the operation history information and the load performance information as well as the guest room information,
The neural network calculation unit cleans the room based on the operation history information in the cleaning load prediction period and the guest room information using the neural network whose neural network coefficients have been determined by the coefficient determination unit. Find the cleaning load when
The cleaning load management device according to claim 1 or 2. further comprising a schedule generation unit that generates schedule information indicating a cleaning schedule for the room based on the cleaning load determined by the neural network calculation unit;
The cleaning load management device according to any one of claims 1 to 3. The schedule generation unit generates the schedule information based on usage status information indicating whether the user is staying in the room.
The cleaning load management device according to claim 4. The neural network calculation unit calculates cleaning for the room based on operation history information for a period from the beginning of the cleaning load prediction period to a point in time past the end of the cleaning load prediction period by a preset margin period. Find the cleaning load when cleaning,
The cleaning load management device according to any one of claims 1 to 5. an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in the room;
A predetermined method for determining the degree of cleaning load required to clean the room with respect to the operation history of the equipment as the degree of work load required when cleaning the room after the user of the room stays in the room. a neural network storage unit that stores information regarding the neural network having the number of nodes and the number of layers determined;
a coefficient acquisition unit that acquires neural network coefficients of the neural network;
The user stays in the room using a neural network set to the neural network coefficients acquired by the coefficient acquisition unit, which is constructed using information regarding the neural network stored in the neural network storage unit. a neural network calculation unit that calculates a cleaning load when cleaning the room from operation history information in a cleaning load prediction period that includes at least a part of the period in which the room is cleaned;
Cleaning load management device. an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in the room;
a cleaning load acquisition unit that acquires load performance information indicating a cleaning load that is a workload required for cleaning the room;
a coefficient determination unit that determines a neural network coefficient of a neural network for determining a cleaning load when cleaning the room after a user stays in the room, based on the operation history information and the load performance information; ,
Using the neural network for which the neural network coefficients have been determined by the coefficient determining unit, the neural network of the room is determined based on operation history information during a cleaning load prediction target period that includes at least a portion of the period during which the user stayed in the room. comprising a neural network calculation unit that calculates a cleaning load when cleaning;
Cleaning load management system. a step in which the cleaning load management system acquires operation history information indicating the operation history of each of at least one device installed in the room;
a step in which the cleaning load management system acquires load performance information indicating a cleaning load that is a workload required for cleaning the room;
Neural network coefficients of a neural network for the cleaning load management system to determine a cleaning load when cleaning the room after the user stays in the room based on the operation history information and the load performance information a step of determining
The cleaning load management system uses the neural network for which the neural network coefficient has been determined to calculate the cleaning load from operation history information in a cleaning load prediction period that includes at least a part of the period during which the user stayed in the room. a step of determining a cleaning load when cleaning a room;
Cleaning load management method. computer,
an operation history acquisition unit that acquires operation history information indicating the operation history of each of at least one device installed in the room;
a cleaning load acquisition unit that acquires load performance information indicating a cleaning load that is a workload required for cleaning the room;
a coefficient determination unit that determines a neural network coefficient of a neural network for determining a cleaning load when cleaning the room after a user stays in the room, based on the operation history information and the load performance information;
Using the neural network for which the neural network coefficients have been determined by the coefficient determining unit, the neural network of the room is determined based on operation history information during a cleaning load prediction target period that includes at least a portion of the period during which the user stayed in the room. Neural network calculation unit that calculates the cleaning load when cleaning;
A program to function as

Images