JP7489653B2 - Presentation method and presentation system - Google Patents

Description

The present disclosure relates to a presentation method and a presentation system.

There is technology that predicts the time it will take to dry laundry using a dryer (see, for example, Patent Document 1).

JP 2021-154049 A

The present disclosure provides a presentation method that appropriately indicates when a device will end processing.

The presentation method in the present disclosure is a presentation method for presenting the timing at which an apparatus ends a process, the method comprising: inputting a parameter value of the apparatus related to a first process, which is the process; inputting parameter values related to each of a plurality of second processes, which are the process, executed by the apparatus in the past, into a prediction model having an input of a parameter value of the apparatus related to a first process, which is the process, and outputting the timing at which the apparatus ends the first process; obtaining a first predicted value of the timing at which the apparatus ends each of a plurality of the second processes; calculating a first reliability of the first predicted value obtained for each of the plurality of the second processes; obtaining an actual measured value of the timing at which the apparatus ends each of a plurality of the second processes; determining a threshold value for the first reliability using the first predicted value, the first reliability, and the actual measured value for each of the plurality of the second processes; and inputting a parameter value related to a third process, which is the process, into the prediction model, which outputs the prediction model; a second predicted value indicating a timing at which the third processing will be terminated is obtained, a second reliability of the second predicted value is calculated, and a presentation control is performed to present the second predicted value, while a notification control is performed to give a notification when the second reliability is less than the threshold value, and in determining the threshold value, the threshold value is determined such that at least one of a first condition and a second condition is satisfied, the first condition is a condition for a first value that is a proportion of one or more second processings in which a difference between the first predicted value and the actual measured value is less than a tolerance value among N times of the second processing in which the first reliability is lower than the threshold value among a plurality of times of the second processing, where N is an integer of 1 or more, and the second condition is a condition for a second value that is a proportion of one or more second processings in which the first reliability is higher than the threshold value among M times of the second processing in which the difference between the first predicted value and the actual measured value is equal to or greater than the tolerance value, where M is an integer of 1 or more.

The presentation method disclosed herein can appropriately present the timing at which the device will finish processing.

FIG. 1 is a schematic diagram showing a configuration of a presentation system according to an embodiment. FIG. 2 is a block diagram showing the functions of a dryer according to an embodiment. FIG. 2 is a block diagram showing functions of a terminal according to the embodiment. FIG. 2 is a block diagram showing the functions of a server according to the embodiment. FIG. 4 is an explanatory diagram showing a change in learning degree according to the embodiment. FIG. 11 is an explanatory diagram showing a transition of a degree of change over time in the embodiment. FIG. 11 is an explanatory diagram showing a transition of reliability in the embodiment. FIG. 11 is an explanatory diagram showing a transition of unreliability in the embodiment. 11 is an explanatory diagram showing whether or not a notification is to be made for each threshold candidate in the embodiment. FIG. 10 is an explanatory diagram showing a false alarm rate and a failure to report rate for each threshold candidate according to an embodiment; FIG. 4 is a flow chart showing the processing of the presentation system in the embodiment. 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; 1 is an explanatory diagram showing an example of an image displayed by a presentation system according to an embodiment; FIG. 11 is a schematic diagram illustrating a configuration of a presentation system according to a first modified example of the embodiment. FIG. 11 is a schematic diagram showing a configuration of a presentation system according to a second modified example of the embodiment.

The inventor has discovered that the presentation method described in the "Background Art" section has the following problems:

There is a technology that predicts the time required for a dryer to dry laundry (see, for example, Patent Document 1). The time prediction is performed, for example, by inputting information such as the pre-drying temperature into a trained model (also called a machine learning model) generated by machine learning.

In general, predictions using machine learning can, in principle, provide prediction times that are adapted to the circumstances of each device.

However, machine learning predictions are not always accurate enough to cover the relatively long lifecycle of an equipment. Low accuracy predictions can result in a different time than the equipment actually needs to dry.

Factors that can cause the accuracy of predictions to decrease include, for example, progress in learning or changes in the equipment over time.

For example, when the input and output information about the device used as training data is not available in the amount necessary for sufficient learning, the learning will be insufficient and the accuracy of predictions will decrease.

In addition, for example, when the equipment changes over time (e.g., the operation of the hardware used by the equipment for processing deteriorates or the sensitivity of the sensors decreases), the accuracy of the predictions decreases.

Therefore, if it were possible to notify the user when a prediction is presented that the prediction has low accuracy, it would be useful for the user to understand how valid the prediction is or how much they can trust the prediction.

The present disclosure provides a presentation method that appropriately indicates when a device will end processing.

A presentation method according to one aspect of the present disclosure is a presentation method for presenting the timing at which an apparatus ends a process, comprising the steps of: inputting a parameter value of the apparatus related to a first process, which is the process; inputting parameter values related to each of a plurality of second processes, which are the process, executed by the apparatus in the past, into a prediction model having an input of a parameter value of the apparatus related to a first process, which is the process, and outputting the timing at which the apparatus ends the first process; obtaining a first predicted value of the timing at which the apparatus ends each of a plurality of the second processes; calculating a first reliability of the first predicted value obtained for each of the plurality of the second processes; obtaining an actual measurement value of the timing at which the apparatus ends each of a plurality of the second processes; determining a threshold value for the first reliability using the first predicted value, the first reliability, and the actual measurement value for each of the plurality of the second processes; and inputting a parameter value related to a third process, which is the process, into the prediction model, which outputs the prediction model; a second predicted value indicating a timing when a device will terminate the third processing, a second reliability of the second predicted value, a presentation control for presenting the second predicted value, and a notification control for issuing a notification when the second reliability is less than the threshold, the determination of the threshold value includes determining the threshold value such that at least one of a first condition and a second condition is satisfied, the first condition is a condition for a first value that is a proportion of one or more second processings in which a difference between the first predicted value and the actual measured value is less than a tolerance value among N times of the second processing in which the first reliability is lower than the threshold value among a plurality of times of the second processing, N is an integer equal to or greater than 1, and the second condition is a condition for a second value that is a proportion of one or more second processings in which the first reliability is higher than the threshold value among M times of the second processing in which the difference between the first predicted value and the actual measured value is equal to or greater than the tolerance value, M is an integer equal to or greater than 1.

According to the above aspect, the presentation method presents a predicted value of the timing when the device will end processing from the parameter values of the device using a machine learning model, and notifies the user if the reliability of the predicted value is relatively low. The reliability indicates the degree to which the predicted value is valid, in other words, the degree to which the predicted value is reliable. The user visually recognizes the presented predicted value of the end timing, and if a notification is given, the user can understand that the reliability of the predicted value is relatively low. In this way, the presentation method can appropriately present the timing when the device will end processing.

Furthermore, the first condition may include a condition that the first value is equal to or less than a predetermined value, or the second condition may include a condition that the second value is equal to or less than a predetermined value.

According to the above aspect, the presentation method determines a threshold value used to judge the reliability of the predicted value of the end timing more easily by using the condition that the first value or the second value is equal to or less than a predetermined value. Therefore, the presentation method can appropriately present the timing at which the device will end processing by easily determining the reliability threshold value.

The first condition may include a condition that the first value is the minimum value among multiple first value candidates calculated using multiple threshold value candidates, or the second condition may include a condition that the second value is the minimum value among multiple second value candidates calculated using multiple threshold value candidates.

According to the above aspect, the presentation method determines a threshold value used to judge the reliability of the predicted value of the end timing more easily by using the condition that the first value or the second value is a minimum value. Therefore, the presentation method can appropriately present the timing at which the device will end processing by easily determining the reliability threshold value.

In addition, the presentation control may include control to display the second predicted value on a display screen, and the notification control may include control to change the display manner of the second predicted value.

According to the above aspect, the presentation method notifies the user by changing the display mode of the predicted value displayed on the display screen. By visually checking the change in the display mode of the predicted value, the user can easily understand that the reliability of the predicted value is relatively low. Thus, the presentation method can appropriately present the timing for the device to end processing by notifying the user of the decrease in reliability by changing the display mode.

In addition, when determining the threshold value, an instruction to raise or lower the determined threshold value may be received from a user, and the determined threshold value may be raised or lowered in accordance with the received instruction, and the notification control may be performed using the threshold value after it has been raised or lowered.

According to the above aspect, the presentation method adjusts the threshold value used to determine the reliability of the predicted value of the end timing based on an instruction received from a user, so that notifications can be provided at an amount or frequency according to the instruction. The user can receive notifications at an amount or frequency according to the instruction. Thus, the presentation method can appropriately present the timing at which the device will end processing using an appropriate amount or frequency of notifications.

In addition, when determining the threshold value, an instruction to set the threshold value may be received from a user, and the threshold value may be set in accordance with the received instruction, and the notification control may use the threshold value after it has been set.

According to the above aspect, the presentation method sets the threshold value used to determine the reliability of the predicted value of the end timing based on an instruction received from a user, so that notifications can be provided at an amount or frequency according to the instruction. The user can receive notifications at an amount or frequency according to the instruction. Thus, the presentation method can appropriately present the timing at which the device will end processing using an appropriate amount or frequency of notifications.

Furthermore, when obtaining the first reliability, the accuracy of the sensor that obtained the parameter values of the equipment related to each of the second processes may be obtained, and the higher the accuracy of the obtained sensor, the higher the first reliability may be obtained, and when obtaining the second reliability, the accuracy of the sensor that obtained the parameter values of the equipment related to the third process may be obtained, and the higher the accuracy of the obtained sensor, the higher the second reliability may be obtained.

According to the above aspect, the method of presenting the parameter easily obtains the reliability according to the accuracy of the sensor that obtains the parameter value. Therefore, the method of presenting the parameter can appropriately present the timing at which the device ends the process.

The device may also be a dryer, and the process may be a process in which the dryer dries laundry.

According to the above aspect, the above presentation method can appropriately present the timing when the dryer will finish the process of drying laundry.

A presentation system according to one aspect of the present disclosure is a presentation system that presents the timing at which an apparatus ends a process, and includes a prediction unit, a reliability calculation unit, a threshold determination unit, and a presentation control unit. The prediction unit inputs a parameter value of the apparatus related to a first process, which is the process, and obtains a first predicted value of the timing at which the apparatus ends each of the multiple second processes, which is output by inputting parameter values related to each of the multiple second processes that the apparatus has previously performed, into a prediction model that outputs the timing at which the apparatus ends the first process. The reliability calculation unit calculates a first reliability of the first predicted value obtained for each of the multiple second processes. The threshold determination unit obtains an actual measurement value of the timing at which the apparatus ends each of the multiple second processes, and determines a threshold for the first reliability using the first predicted value, the first reliability, and the actual measurement value for each of the multiple second processes. The prediction unit inputs a parameter value of the apparatus related to a third process, which is the process, into the prediction model. a presentation system including: a second predicted value indicating a timing at which the device will terminate the third processing, the second predicted value being output by inputting a parameter value; a reliability calculation unit calculating a second reliability of the second predicted value; a presentation control unit performing presentation control to present the second predicted value and performing notification control to give a notification when the second reliability is less than the threshold; determining the threshold value determines the threshold value at which at least one of a first condition and a second condition is satisfied; the first condition is a condition for a first value which is a proportion of one or more second processings in which a difference between the first predicted value and the actual measured value is less than a tolerance value among N second processings in which the first reliability is lower than the threshold value among a plurality of second processings, where N is an integer equal to or greater than 1; and the second condition is a condition for a second value which is a proportion of one or more second processings in which the first reliability is higher than the threshold value among M second processings in which the difference between the first predicted value and the actual measured value is equal to or greater than the tolerance value, where M is an integer equal to or greater than 1.

According to the above aspect, the same effect as the above presentation method is achieved.

Below, the embodiments will be described in detail, with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of matters that are already well known or duplicate explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to make it easier for those skilled in the art to understand.

The inventor(s) provide the accompanying drawings and the following description to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

(Embodiment)
In this embodiment, a presentation method and a presentation system for appropriately presenting the timing at which a device ends a process will be described.

Figure 1 is a schematic diagram showing the configuration of the presentation system 1 in this embodiment.

1, the presentation system 1 includes a server 10, dryers 20A, 20B, and 20C (also referred to as dryer 20A, etc.), and terminals 30A, 30B, and 30C (also referred to as terminal 30A, etc.). The devices included in the presentation system 1 are connected to each other so as to be able to communicate with each other via a network N.

The server 10 calculates a predicted value of the timing (also called the end timing) when the dryer 20A or the like ends processing and the reliability of the predicted value. The end timing is predicted, for example, by a method using machine learning, and this case will be described as an example, but is not limited to this. The server 10 provides the calculated predicted value and reliability to the terminal 30A.

Here, the terminal to which the server 10 provides the predicted value and reliability of the dryer 20A, etc. may be a terminal among the terminals 30A, etc. that is associated with the dryer 20A, etc. The association between the dryer 20A, etc. and the terminal 30A, etc. is performed by the server 10.

The dryer 20A is a device that executes a process. The dryer 20A is, for example, a dryer (which can also be called a washing machine with a dryer) that dries laundry such as selected clothes, and this case will be described as an example, but is not limited to this. The process performed by the dryer 20A is a process of drying laundry (also called a drying process). In addition to the above, the dryer 20A may be a device such as a freezer or an air conditioner. If the target device is a freezer, the process is an ice-making process. If the target device is an air conditioner, the process is a heating process (more specifically, for example, a heating process that raises the temperature in the bedroom to 25°C).

Dryers 20B and 20C are each similar devices to dryer 20A and operate independently of dryer 20A.

Terminal 30A presents information. Terminal 30A presents a predicted value of the end timing calculated by server 10. Terminal 30A also notifies when the reliability of the predicted value is less than a threshold value.

The presentation may be, for example, displaying an image including the information on a display screen of the terminal 30A, but is not limited to this, and may be output as sound. The notification may be, for example, displaying an image including the information on a display screen of the terminal 30A, or changing the display mode of the information displayed on the display screen of the terminal 30A, but is not limited to this, and may be output as sound or changing the output mode of the sound. The change in the output mode of the sound may be, for example, a change in the volume, a change in the playback speed, a change in whether or not an effect is applied, a change in the effect applied, etc.

Terminals 30B and 30C are each devices similar to terminal 30A and operate independently of terminal 30A.

Here, an example is described in which there are three dryers 20A, etc. and three terminals 30A, etc., but this is not limited to this. There may be any number of dryers 20A, etc. or terminals 30A, etc. Also, the number of dryers 20A, etc. and the number of terminals 30A, etc. may differ.

Network N may be composed of any communication line or network, including, for example, the Internet, a mobile phone carrier network, an Internet provider access network, or a public access network.

The following describes in detail the functions of the dryers 20A, etc., the terminals 30A, etc., and the server 10. Here, the description focuses on the dryer 20A among the dryers 20A, etc., but the same description also applies to the dryers 20B and 20C.

Figure 2 is a block diagram showing the functions of dryer 20A in this embodiment. Note that dryers 20B and 20C have similar functions, so the explanation of dryers 20B and 20C is omitted.

As shown in FIG. 2, the dryer 20A has, as functional units, a communication unit 21, a processing unit 22, and a display unit 23.

The communication unit 21 is a communication interface that can be connected to the network N. The communication unit 21 transmits and receives communication packets between the functional unit of the dryer 20A and other devices when the functional unit communicates with the other devices through the network N. The communication unit 21 is, for example, a Wi-Fi (registered trademark) communication interface.

The processing unit 22 controls the drying process, which is a process performed by the dryer 20A. Specifically, the processing unit 22 has a tub (generally also used as a washing tub) into which the laundry to be processed is put, a heat pump that takes in air in the tub to generate hot air and sends it into the tub, and a control unit that controls the operation of the tub and the heat pump. The processing unit 22 also has one or more sensors that detect information related to the drying process. The one or more sensors include, for example, a weight sensor that detects the weight of the laundry in the tub, a temperature sensor that detects the temperature of the air taken in from the tub to the heat pump (also called the intake temperature), or a temperature sensor that detects the temperature of the air sent from the heat pump to the tub (also called the discharge temperature). The processing unit 22 determines the drying time, which is the length of time for the drying process. The above drying time is determined, for example, by referring to the drying time determined for each weight of the laundry at the time when drying is started, and selecting a drying time that matches the weight of the laundry to be dried and detected by the weight sensor. The processing unit 22 starts the drying process, and ends the drying process when the drying time has elapsed since the start of the drying process. The processing section 22 may successively adjust the drying time determined at the start of the drying process based on the temperature detected by the temperature sensor as the drying progresses.

In addition, the processing unit 22 transmits information detected by one or more sensors to the server 10 in order for the server 10 to calculate a predicted value of the end timing. The information detected by the one or more sensors includes, for example, the weight of the laundry detected by a weight sensor, the suction temperature or discharge temperature detected by a temperature sensor, etc.

The display unit 23 is a display device that displays information. For example, the display unit 23 displays information including a predicted value of the end timing obtained from the server 10 as an image. The display unit 23 also notifies by displaying the image or changing the display mode. The display unit 23 may be a speaker that outputs the information as sound. The display unit 23 may be a presentation unit having at least one of the display device and the speaker.

Figure 3 is a block diagram showing the functions of terminal 30A in this embodiment. Terminals 30B and 30C also have similar functions, so descriptions of terminals 30B and 30C are omitted.

As shown in FIG. 3, the terminal 30A has, as functional units, a communication unit 31, a control unit 32, and a display unit 33.

The communication unit 31 is a communication interface that can be connected to the network N. The communication unit 31 transmits and receives communication packets between the terminal 30A and other devices when the functional unit of the terminal 30A communicates with the other devices through the network N. The communication unit 31 is, for example, a Wi-Fi communication interface or a mobile phone network communication interface.

The control unit 32 controls the operation of the terminal 30A. The control unit 32 has a processor, memory, etc., and controls the operation of the terminal 30A by the processor executing a predetermined program using the memory.

The display unit 33 is a display device that displays information. For example, the display unit 33 displays information including a predicted value of the end timing obtained from the server 10 as an image. The display unit 33 notifies by displaying the image or changing the display mode. The display unit 33 may be a speaker that outputs the information as sound. The display unit 33 may also be a presentation unit having at least one of the display device and the speaker.

Figure 4 is a block diagram showing the functions of server 10 in this embodiment.

4, the server 10 includes, as functional units, a communication unit 11, a training unit 12, a prediction unit 13, a reliability calculation unit 14, a threshold determination unit 15, and a display control unit 16. The functional units of the server 10 can be realized by a processor of the server 10 executing a predetermined program using a memory.

The communication unit 11 is a communication interface that can be connected to the network N. The communication unit 11 transmits and receives communication packets between the functional unit of the server 10 and other devices when the functional unit communicates with the other devices through the network N. The communication unit 11 is, for example, an Ethernet (registered trademark) communication interface.

The training unit 12 has a prediction model, which is a machine learning model that predicts the end timing, and trains the prediction model. The prediction model is a model in which the parameter values of the dryer 20A related to the drying process (corresponding to the first process) are input, and the end timing at which the dryer 20A ends the drying process is output. The prediction model is constructed by training by machine learning using teacher data in which the parameter values of the dryer 20A related to the drying process are input, and the timing at which the dryer 20A ends the drying process is output. The teacher data may be teacher data in which the parameter values of the dryer 20A related to the actual drying process are input data, and the timing at which the dryer 20A ends the drying process is output data, or it may be teacher data in which the input data and the output data are for equipment of the same type as the dryer 20A. The prediction model is, for example, a neural network model.

The prediction unit 13 uses the prediction model to obtain a predicted value of the end timing of the drying process performed by the dryer 20A. Specifically, the prediction unit 13 obtains information detected by one or more sensors possessed by the dryer 20A for each of the multiple drying processes (corresponding to the second process) performed in the past by the dryer 20A, and obtains a predicted value of the end timing (also called a first predicted value) that is output by inputting the obtained information into the prediction model as a parameter value. In addition, the prediction unit 13 obtains information detected by one or more sensors possessed by the dryer 20A for the drying process (corresponding to the third process) performed by the dryer 20A, and obtains a predicted value of the end timing (also called a second predicted value) that is output by inputting the obtained information into the prediction model as a parameter value.

The reliability calculation unit 14 calculates the reliability (first reliability) of the end timing, which is the first predicted value acquired by the prediction unit 13. The reliability calculation unit 14 also calculates the reliability (second reliability) of the end timing, which is the second predicted value acquired by the prediction unit 13. The reliability indicates the degree to which the predicted value is valid, in other words, the degree to which the predicted value can be trusted.

The reliability calculation unit 14, for example, obtains the accuracy of the sensor that obtained the parameter values of the dryer 20A related to the drying process, and obtains a higher reliability the higher the accuracy of the obtained sensor.

In addition, when calculating the reliability, the reliability calculation unit 14 can obtain, for example, a learning degree, which is an index showing the progress of learning of the prediction model, and a change rate over time, which is an index showing the degree of change over time of the dryer 20A, and calculate the reliability using at least one of the obtained learning degree and change rate over time. A method for calculating the reliability using the learning degree and change rate over time will be described later.

The threshold determination unit 15 determines a threshold for the first reliability using the first predicted value, the first reliability, and the actual measured value for each of the multiple second processes. In determining the threshold, the threshold determination unit 15 determines a threshold that satisfies at least one of the following first condition and the following second condition:

The first condition is a condition regarding the false alarm rate. The false alarm rate is an index used in a confusion matrix, and indicates the proportion of cases in which it is determined that notification should not be made based on the difference between the predicted value and the actual value of the end timing, among cases in which it is determined that notification should be made because the reliability is lower than a threshold. More specifically, the first condition is a condition regarding the false alarm rate (corresponding to the first value), which is the proportion of one or more second processes in which the difference between the first predicted value and the actual value is less than the tolerance, among N second processes in which the reliability is lower than a threshold, among multiple second processes. Here, N is an integer equal to or greater than 1.

The first condition may include a condition that the false alarm rate is equal to or less than a predetermined value. The first condition may also include a condition that the false alarm rate is the minimum value among multiple false alarm rate candidates calculated using multiple threshold candidate values.

The second condition is a condition regarding the false alarm rate. The false alarm rate is an index used in a confusion matrix, and indicates the proportion of cases in which a notification is not issued because the reliability is higher than a threshold among cases in which it is determined that a notification should be issued based on the difference between the predicted value and the actual measured value of the end timing. More specifically, the second condition is a condition regarding the false alarm rate (corresponding to the second value), which is the proportion of at least one second process whose reliability is higher than a threshold among M second processes in which the difference between the first predicted value and the actual measured value among multiple second processes is equal to or greater than the tolerance. Here, M is an integer equal to or greater than 1.

The second condition may include a condition that the failure rate is equal to or less than a predetermined value. The second condition may also include a condition that the failure rate is the minimum value among multiple failure rate candidates calculated using multiple threshold candidate values.

For example, the first condition can be "the false alarm rate is 0.3 or less" and the second condition can be "the missed alarm rate is 0.3 or less."

For example, the first condition can be "the false alarm rate is 0.5 or less" and the second condition can be "the smallest value among the candidate false alarm rates."

The false alarm rate and false alarm rate can be expressed as follows using precision and recall, which are commonly used in confusion matrices:

False alarm rate = 1-precision
Failure rate = 1-recall

In addition to the above, a method of determining an appropriate threshold using a confusion matrix may also be adopted.

The display control unit 16 performs display control to display presentation information including the second predicted value acquired by the prediction unit 13. The display control corresponds to presentation control to present the presentation information. In addition to the presentation control, the display control unit 16 also performs notification control to notify when the second reliability is less than a threshold value. The display control unit 16 corresponds to a presentation control unit.

Specifically, in the display control, the display control unit 16 transmits the presentation information to the terminal 30A and controls the terminal 30A to display the presentation information on the display unit 33. In addition, in the notification control, the display control unit 16 controls the terminal 30A to change the display mode.

The display control unit 16 has correspondence information that associates a dryer with a terminal that presents and notifies the user of the predicted value and reliability of the end timing for the dryer, and selects a terminal that presents the presented information by referring to the correspondence information. In the correspondence information, dryers 20A, 20B, and 20C are associated with terminals 30A, 30B, and 30C, respectively. For example, the terminal that corresponds to dryer 20A may be a terminal owned by a user who owns dryer 20A.

In addition, the display control unit 16 may transmit the presentation information to the dryer 20A and control the dryer 20A to display the presentation information on the display unit 23.

When determining the threshold, the threshold determination unit 15 may receive an instruction from the user to raise or lower the determined threshold, and raise or lower the determined threshold in accordance with the received instruction. In this case, the display control unit 16 performs notification control using the raised or lowered threshold in notification control.

When determining the threshold value, the threshold value determination unit 15 may receive an instruction to set the threshold value from the user and set the threshold value according to the received instruction. In this case, the display control unit 16 performs notification control using the threshold value after it has been set.

Next, we will explain reliability.

Here, we explain the reliability calculated using the degree of learning and the degree of change over time.

Figure 5 is an explanatory diagram showing the progress of learning level α in this embodiment.

In Figure 5, the horizontal axis indicates the number of processing times, and the vertical axis indicates the learning degree α. The learning degree α is an index that increases as the learning of the predictive model progresses. The learning degree α shown in Figure 5 is an index that indicates the progress of the learning of the predictive model in the range of 0 to 1, and will be described as an example, but the range of the learning degree α is not limited to the above.

As shown in Figure 5, the learning degree α has a relatively low value when the number of processing times is relatively small, and has a higher value as the number of processing times increases. For example, the learning degree α2 at the processing time P2 is higher than the learning degree α1 at the processing time P1.

Furthermore, after a certain number of processing times, the learning level α takes a value of 1 or close to 1, and thereafter continues to take a value of 1 or close to 1. For example, at processing time P3, the learning level α is approximately 1, and even when the processing time is P3 or more, the learning level α remains approximately 1.

The learning level α is calculated by the reliability calculation unit 14 using at least one of the following pieces of information, for example:

(a) The value obtained by dividing the training error by the validation error in the cross-validation of the predictive model. (b) The number of data used to train the predictive model. (c) The convergence rate of the prediction error of the first predicted value in multiple processings performed by the device.

Figure 6 is an explanatory diagram showing the progression of change over time in this embodiment.

In Figure 6, the horizontal axis indicates the time elapsed since the time of manufacture of the dryer 20A, and the vertical axis indicates the degree of change over time β. The degree of change over time β is an index indicating the degree of change over time since the time of manufacture of the dryer 20A, which is 0 at the time of manufacture and increases as the time elapsed after the time of manufacture increases.

As shown in Figure 6, the degree of change over time β has a relatively low value when the elapsed time is relatively short, and has a higher value as the elapsed time is longer. For example, the degree of change over time β3 at elapsed time T3 is higher than the degree of change over time β2 at elapsed time T2.

The degree of change over time β is calculated by the reliability calculation unit 14 using at least one of the following pieces of information, for example. Here, N is an integer equal to or greater than 2.

(d) The average value of the prediction errors of the first predicted value in the most recent N processes among the multiple processes performed by the dryer 20A. (e) The value obtained by dividing the above (d) by the average value of the prediction errors of the first predicted value in processes performed by multiple devices of the same type as the dryer 20A. (f) The standard deviation of the timing at which the dryer 20A completed the process in the most recent N processes among the multiple processes performed by the dryer 20A.

Figure 7 is an explanatory diagram showing the progress of reliability in this embodiment.

7, the horizontal axis indicates the time elapsed since the manufacturing time of the dryer 20A, and the vertical axis indicates the reliability δ. The reliability δ is an index indicating the degree to which the predicted value of the end timing at which the dryer 20A ends the process can be trusted, and is an index that varies in the range from 0 to 1.

The reliability δ is calculated using the learning level α and the degree of change over time β, and more specifically, the higher the learning level α, the higher the reliability δ, or the higher the degree of change over time β, the lower the reliability δ.

For ease of explanation, hereafter, unreliability ε will be used as an index of reliability. Unreliability is an index whose magnitude relationship is the reverse of that of reliability δ. Unreliability ε is an index that ranges from 0 to 1, and the lower the unreliability ε, the higher the reliability of the predicted value of the end timing at which dryer 20A will end processing (see Equation 1 below).

ε = 1 - δ (Equation 1)

Next, we will explain how to determine the threshold. The threshold determination unit 15 determines the threshold by selecting one threshold from multiple threshold candidates (also called threshold candidates).

Figure 9 is an explanatory diagram showing whether or not notification is appropriate for each threshold candidate in this embodiment.

Figure 9 shows the drying time, predicted drying time, difference between the drying time and predicted drying time (also simply called "difference"), notification flag, and unreliability for each of the 10 drying processes previously performed by dryer 20A. Figure 9 also shows the unreliability flag and whether or not a notification was issued for each of the 10 drying processes, assuming three potential threshold values.

The drying time is the time required for the drying process calculated at the time when the dryer 20A starts the drying process. The drying time is an example of an actual measured value of the end timing.

The predicted drying time is a predicted value of the time required for the drying process, predicted using the parameter values of the dryer 20A at the time when the dryer 20A starts the drying process. The predicted drying time is an example of a predicted value of the end timing.

The difference between the drying time and the predicted drying time is the predicted drying time minus the drying time.

The notification flag is a flag that indicates that a notification should be made based on the difference between the drying time and the predicted drying time for each of the 10 drying processes. The notification flag is set to 1 (i.e., notification should be made) if the absolute value of the difference exceeds the tolerance, and is set to 0 (i.e., notification should not be made) if the absolute value of the difference is equal to or less than the tolerance. The tolerance is, for example, 15 minutes, and this case will be explained as an example. It does not matter whether a notification was actually made for a drying process for which the notification flag is 1.

The unreliability is the unreliability of the predicted drying time. The unreliability can be calculated using the parameter values of the dryer 20A at the time the dryer 20A starts the drying process.

For example, for drying process #1, the drying time is 90 minutes, the predicted drying time is 95 minutes, and the difference is 5 minutes. The notification flag is 0 because the absolute value of the difference, 5 minutes, is less than the tolerance of 15 minutes. The unreliability is 10.

For example, in drying process #2, the drying time is 100 minutes and the predicted drying time is 120 minutes, with a difference of 20 minutes. The notification flag is 1 because the absolute value of the difference, 20 minutes, exceeds the tolerance of 15 minutes. The unreliability is 20.

For example, in drying process #3, the drying time is 80 minutes and the predicted drying time is 70 minutes, resulting in a difference of -10 minutes. The notification flag is 0 because the absolute value of the difference, 10 minutes, is less than the tolerance of 15 minutes. The unreliability is 25.

Next, the appropriateness of notification for each threshold candidate will be explained. Here, an example will be explained in which there are three threshold candidate thresholds: 20, 50, and 80. Note that the threshold candidates are not limited to the above three, and multiple threshold candidates within the range from the minimum to maximum unreliability values may be used. For example, the threshold candidates may be all integer values within the range from the minimum value of 0 to the maximum value of 100.

The unreliability flag is a flag that indicates whether the unreliability is equal to or greater than the threshold value when the threshold candidate is used as the threshold value for each of the 10 drying processes. The unreliability flag is set to 1 if the unreliability in the drying process is equal to or greater than the threshold value, and 0 if not.

The appropriateness of notification is a flag indicating whether the threshold determination unit 15 has determined that a notification should be made for a drying process whose unreliability flag is 1. The appropriateness of notification indicates whether it is appropriate to make a notification for a drying process whose unreliability flag is 1, based on the difference. The appropriateness of notification is set to 1 when the notification flag is 1 for a drying process whose unreliability flag is 1, and is set to 0 otherwise. Note that the appropriateness of notification is not defined for a drying process whose unreliability flag is 0, and is indicated as "-" in the figure.

For example, in drying process #1, when the candidate threshold value 20 is used as the threshold value, the unreliability flag is 0. This is because the unreliability level of 10 is less than the candidate threshold value of 20. Also, in drying process #1, when the candidate threshold values of 50 or 80 are used as the threshold value, the result is the same as when the candidate threshold value of 20 is used as the threshold value.

For example, in drying process #2, when the candidate threshold value of 20 is used as the threshold value, the unreliability flag is 1. This is because the unreliability of 20 is greater than or equal to the candidate threshold value of 20. Also, in drying process #2, the unreliability flag is 1 and the notification flag is 1, so the notification is correct or not correct.

In addition, in drying process #2, the unreliability flag is 0 when the threshold candidate values of 50 or 80 are used as the threshold value. This is because the unreliability level of 20 is less than the threshold candidate values of 50 or 80.

For example, in drying process #3, when the candidate threshold value of 20 is used as the threshold value, the unreliability flag is 1. This is because the unreliability of 25 is greater than or equal to the candidate threshold value of 20. Also, in drying process #3, the unreliability flag is 1 and the notification flag is 0, so the validity of the notification is 0.

In addition, in drying process #3, the unreliability flag is 0 when the threshold candidate values of 50 or 80 are used as the threshold value. This is because the unreliability level of 25 is less than the threshold candidate values of 50 or 80.

Figure 10 is an explanatory diagram showing the false alarm rate and missed alarm rate for each threshold candidate in this embodiment.

Figure 10 shows, for each of several threshold candidate values, the number of processes that trigger a notification, the number of processes that trigger a notification, the false alarm rate, and the failure to notify rate.

The number of processes for which notification is to be sent indicates the number of drying processes for which notification should be sent when the threshold candidate is used as the threshold, and is the number of drying processes for which the unreliable flag is 1 when assuming the threshold candidate shown in Figure 9.

The number of processes for which a notification is appropriate is the number of drying processes for which the threshold determination unit 15 has determined that a notification should be issued, and for which the appropriateness of the notification is 1 when considering the threshold candidate values shown in Figure 9.

The false alarm rate indicates the false alarm rate when the candidate threshold is used as the threshold, and is calculated using the following formula (2).

False alarm rate = ((number of processes that notify) - (number of processes that are appropriate for notification)) / (number of processes that notify) (Equation 2)

For example, the false alarm rate for the candidate threshold of 20 is calculated to be (9-4)/9 ≒ 0.56.

The failure rate indicates the failure rate when the threshold candidate is used as the threshold, and is calculated using the following formula (3). The number of processes for which notifications should be sent is the number of drying processes whose notification flag is 1, which is 4.

Failure rate = ((number of processes that should be notified) - (number of processes for which notification is appropriate)) / (number of processes that should be notified) (Equation 3)

For example, the false positive rate for the candidate threshold of 50 is calculated as (4-3)/4 = 0.25.

Note that the false alarm rate and missed alarm rate for each of the multiple threshold candidate values shown in Figure 10 correspond to multiple false alarm rate and missed alarm rate candidates, respectively.

The processing of the presentation system 1 configured as described above will be explained.

Figure 11 is a flow diagram showing the processing of the presentation system 1 in this embodiment.

In step S101, the prediction unit 13 acquires parameter values for each of the multiple drying processes (corresponding to the second process) that the dryer 20A has previously performed.

In step S102, the prediction unit 13 inputs the parameter values acquired in step S101 into the prediction model to acquire a predicted value of the end timing at which the dryer 20A will end the drying process previously performed. The reliability calculation unit 14 acquires the reliability of the predicted value acquired by the prediction unit 13 in step S101.

In step S103, the threshold determination unit 15 obtains the actual measured value of the end timing at which the dryer 20A ends the drying process previously performed.

In step S104, the threshold determination unit 15 determines a threshold for reliability.

In step S105, the prediction unit 13 obtains parameter values for each drying process (corresponding to the third process) performed by the dryer 20A.

In step S106, the prediction unit 13 inputs the parameter values acquired in step S105 into the prediction model to acquire a predicted value of the end timing for terminating the drying process performed by the dryer 20A. The reliability calculation unit 14 acquires the reliability of the predicted value acquired by the prediction unit 13 in step S106.

In step S107, the display control unit 16 controls the display of the predicted value obtained in step S106.

In step S108, the display control unit 16 determines whether the reliability obtained in step S106 is higher than the threshold determined in step S104. If the reliability is higher than the threshold (Yes in step S108), the process proceeds to step S110. If not (No in step S108), the process proceeds to step S109.

In step S109, the display control unit 16 executes notification control and then terminates the series of processes shown in FIG. 11.

In step S110, the display control unit 16 does not execute notification control (in other words, suppresses execution of notification control), and then terminates the series of processes shown in FIG. 11.

In addition, when the prediction unit 13 acquires the parameter values in step S101, the parameter values may be acquired for drying processes that have been performed multiple times in the past by the dryer 20A and that have the same operating mode as the process to be performed. In this way, the threshold value can be determined based on the drying process of the same operating mode (see step S104), and the validity of the threshold value determined for the drying process of the above operating mode can be improved.

By using the process shown in FIG. 11, the presentation system 1 can appropriately present the timing at which the device will end processing.

Below, we will explain in detail the display by the presentation system 1.

The display by the presentation system 1 is, for example, a display by the display unit 33 provided in the terminal 30A, and this case will be described as an example, but the display may also be a display by the display unit 23 of the dryer 20A.

The terminal 30A receives presentation information including the second predicted value calculated by the server 10 from the server 10, and displays the second predicted value as an image on the display screen. When sending a notification, the terminal 30A further displays a notification on the display screen or changes the display mode in response to the notification.

Figure 12 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in FIG. 12 is an image showing a predicted value of the end timing of the drying process performed by the dryer 20A. Note that FIG. 12 is an image that does not include a notification.

The images shown in FIG. 12 include image 51.

Image 51 displays the predicted end timing of the drying process performed by dryer 20A as the time remaining from the current time. Specifically, image 51 includes a rectangular solid-line frame and the text "1:40 remaining" within the frame, indicating that the predicted end timing of the drying process is 1 hour and 40 minutes from the current time.

A user who is presented with the image shown in Figure 12 can grasp the predicted end timing.

Figure 13 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment. Note that images that are the same as images already described are given the same reference numerals, and detailed descriptions may be omitted.

The image shown in Figure 13 shows a predicted value for the end timing of the drying process performed by dryer 20A, and further shows that the predicted value was obtained by prediction based on AI (Artificial Intelligence).

The images shown in Figure 13 include image 51 (similar to Figure 12) showing a predicted value of the end timing of the drying process performed by dryer 20A, and image 52 showing that the predicted value was obtained by prediction based on AI.

Image 52 specifically includes a circular frame and the character string "AI" contained within the frame.

When a user is presented with the image shown in Figure 13, they can understand that the predicted end timing is obtained through AI-based prediction.

Figure 14 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 14 shows a predicted end timing of the drying process performed by dryer 20A and further includes a notification.

The images shown in FIG. 14 include image 51A showing a predicted value of the end timing of the drying process performed by dryer 20A, and image 52A showing that the predicted value was obtained by prediction based on AI.

Image 51A, like image 51, is an image that displays a predicted value for the end timing of the drying process performed by dryer 20A as the remaining time from the current time. However, the color of the rectangular frame included in image 51A is lighter than the color of the frame included in image 51. The lighter color of the frame is an example of a change in the display mode in response to a notification.

Image 52A, like image 52, is an image indicating that the predicted value was obtained by AI-based prediction. However, the color of the circular frame included in image 52A is lighter than the frame included in image 51. The lighter color of the frame is an example of a change in the display mode in response to a notification.

A user presented with the image shown in Figure 14 can understand the predicted end timing and that the reliability of that predicted value is relatively low.

Figure 15 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 15 shows a predicted end timing of the drying process performed by dryer 20A and further includes a notification.

The images shown in FIG. 15 include image 51B showing a predicted value of the end timing of the drying process performed by dryer 20A, and image 52B showing that the predicted value was obtained by prediction based on AI.

Image 51B, like image 51, is an image that displays a predicted value for the timing of completion of the drying process performed by dryer 20A as the remaining time from the current time. However, the rectangular frame included in image 51B is shown with a dashed line. The dashed line indicates that image 51B is flashing, and is an example of a change in the display mode in response to a notification.

Image 52B, like image 52, is an image indicating that the predicted value was obtained by AI-based prediction. However, the circular frame included in image 52B is indicated by a dashed line. The dashed line indicates that image 51B is flashing, and is an example of a change in the display mode in response to a notification.

A user presented with the image shown in FIG. 15 can understand the predicted end timing and that the reliability of that predicted value is relatively low.

Figure 16 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 16 includes a notification of the predicted end timing of the drying process performed by dryer 20A.

The images shown in FIG. 16 include image 51C, which is placed at a position where an image (e.g., image 51) indicating a predicted value of the end timing of the drying process performed by dryer 20A should be placed, and image 52C, which is placed at a position where an image (e.g., image 52) indicating that the predicted value was obtained by prediction based on AI should be placed.

Image 51C includes a rectangular solid-line frame and the symbol "..." within the frame. This is because the predicted value for the end timing of the drying process has a relatively low reliability, so the symbol is displayed instead of displaying the predicted value. The display of the symbol "..." is an example of a change in the display mode in response to a notification. Note that the symbol "..." may be an animation in which multiple "..."s are displayed and hidden in sequence, causing the "..."s to flow left and right.

Image 52C includes a circular frame and the symbol "..." within the frame. This is because the reliability of the predicted value for the timing of completion of the drying process is relatively low, so the symbol is displayed instead of displaying "AI." The symbol "..." is the same as that described for image 51B above.

A user presented with the image shown in Figure 16 can understand that the reliability of the predicted end timing is relatively low.

Figure 17 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 17 shows a predicted value for the end timing of the drying process performed by dryer 20A and further includes a notification.

The images shown in FIG. 17 include image 51D showing a predicted value of the end timing of the drying process performed by dryer 20A, and image 52D showing that the predicted value was obtained by prediction based on AI.

Image 51D, like image 51, displays a predicted value for the end timing of the drying process performed by dryer 20A as the remaining time from the current time. However, image 51D displays an "X" mark along with the predicted value. The display of the "X" mark is an example of a change in the display mode corresponding to a notification. The "X" mark may be displayed throughout the entire frame as shown in FIG. 17, or may instead be displayed in a portion of the frame.

Image 52D, like image 52, is an image indicating that the predicted value was obtained by prediction based on AI. However, image 52D displays a "?" mark along with the character string "AI." The display of the "?" mark is an example of a change in the display mode corresponding to a notification. Note that the "?" mark may be displayed in a part of the frame as shown in FIG. 17, or alternatively, may be displayed throughout the entire frame.

A user presented with the image shown in Figure 17 can understand the predicted end timing and that the reliability of that predicted value is relatively low.

Figure 18 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 18 shows a predicted end timing of the drying process performed by dryer 20A and further includes a notification.

The images shown in FIG. 18 include image 51 showing a predicted value of the end timing of the drying process performed by dryer 20A, image 52 showing that the predicted value was obtained by AI-based prediction, and image 53 notifying that the reliability of the predicted value is relatively low.

Image 53 is an image that notifies that the reliability of the predicted value is relatively low. Image 53 includes the character string "AI recalculating". The character string "AI recalculating" indirectly indicates that the reliability of the predicted value is relatively low. If a character string that directly indicates that the reliability of the predicted value is relatively low, such as "The reliability of the predicted value is low," is used, there is a risk that the user will be misled into thinking that the reliability of the AI technology itself is low, and this has the effect of avoiding this. Note that the character string may also be a character string that directly indicates that the reliability of the predicted value is relatively low, such as "The reliability of the predicted value is low."

A user presented with the image shown in Figure 18 can understand the predicted end timing and that the reliability of that predicted value is relatively low.

Figure 19 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 19 shows a predicted end timing of the drying process performed by dryer 20A and further includes a notification.

The images shown in FIG. 19 include image 51 showing a predicted value of the end timing of the drying process performed by dryer 20A, image 52 showing that the predicted value was obtained by AI-based prediction, and image 54 notifying that the reliability of the predicted value is relatively low.

Image 54 is an image of an icon displayed in the notification area (so-called status bar) of terminal 30A. It is assumed that a user who sees image 54 performs an operation on terminal 30A to view the notification corresponding to the image, and as a result of that operation, for example, image 53 (see FIG. 18) is displayed.

A user presented with the image shown in Figure 19 can understand the predicted end timing and that the reliability of that predicted value is relatively low.

The following describes the adjustment of the threshold for the reliability of the predicted value. In principle, the threshold for the reliability of the predicted value is calculated by the threshold determination unit 15 as described above, but the threshold calculated by the threshold determination unit 15 can be adjusted in response to a user specification. The following describes the image for receiving the user specification and the adjustment in response to the user's instruction.

Figure 20 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 20 is an image displayed by the presentation system 1 to allow the user to specify a threshold value.

The image shown in Figure 20 includes an image 61 containing a message and an image 62 for receiving a threshold value.

Image 61 includes a message for receiving specifications for adjusting the threshold. The message includes the character string "Adjust recalculation sensitivity", the character string "Adjust the sensitivity for recalculating the remaining drying time" as an explanation of the sensitivity adjustment, and the character string "The higher the sensitivity, the more frequent recalculations will occur and the more notifications will be given" as an explanation of the effect of the adjustment.

Image 62 includes a slider 63 and a handle 64. The right end of slider 63 corresponds to a relatively high threshold, and the left end corresponds to a relatively low threshold. The user can specify the threshold by moving handle 64 left and right along slider 63.

The threshold value may be specified relative to the threshold value calculated by the threshold value determination unit 15, such as increasing or decreasing the threshold value calculated by the threshold value determination unit 15 by a predetermined value. The threshold value may be specified absolutely, such as setting the threshold value to a predetermined value regardless of the threshold value calculated by the threshold value determination unit 15.

When a user is presented with the image shown in Figure 20, he or she can understand what adjusting the recalculation sensitivity means and what effect it has, and can also specify the sensitivity adjustment.

Figure 21 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 21 is an image displayed by the presentation system 1 to allow the user to specify a threshold value.

The images shown in Figure 21 include an image 71 containing a message and an image 72 for receiving a threshold value.

Image 71 includes a message for receiving specifications for adjusting the threshold. The message includes the character string "Adjust frequency of accuracy degradation notifications" and the character string "How many notifications would you like to receive about the remaining drying time?" as a question to the user regarding the frequency.

Image 72 includes a slider 73 and a handle 74. The right end of slider 73 corresponds to receiving notifications relatively frequently ("want to receive more"), and the left end corresponds to receiving notifications relatively frequently ("want to receive less"). Slider 73 can be specified on a five-level scale from 1 to 5. The user can adjust the frequency of notifications by moving handle 74 left and right along slider 73.

The threshold value may be specified relative to the threshold value in the same manner as in FIG. 20, or may be specified absolutely.

When a user is presented with the image shown in Figure 21, they can specify how often they would like to receive notifications.

Figure 22 is an explanatory diagram showing an example of an image displayed by the presentation system 1 in this embodiment.

The image shown in Figure 22 is an image displayed by the presentation system 1 to receive a user's evaluation of the notification.

The images shown in FIG. 22 include an image 81 containing a message and an image 82 for receiving an evaluation of the notification.

Image 81 includes a message for receiving a designation for rating the notification (more specifically, an evaluation of the amount of notification). The message includes the character string "Please rate the notification," the character string "Is the notification of the remaining drying time appropriate?" as a question to the user about the rating, and the character string "The AI will learn and provide an appropriate notification" as an explanation of the effect of specifying the rating.

Image 82 includes a slider 83 and a handle 84. The right end of the slider 83 corresponds to the amount of notifications being less than the user expected ("too few"), and the left end corresponds to the amount of notifications being more than the user expected ("too many"). The slider 83 is designated on a five-level scale. The user can rate the notifications by moving the handle 84 left and right along the slider 83.

The threshold value may be specified relative to the threshold value in the same manner as in FIG. 20, or may be specified absolutely.

A user presented with the image shown in FIG. 22 can rate the notification.

In addition, some or all of the functional parts of the server 10 in the presentation system 1 may be provided in the dryer 20A or the terminal 30A.

(First Modification of the Embodiment)
In this modification, a presentation method and system for appropriately presenting the timing when a device ends a process, which is different from the above embodiment, will be described.

Figure 23 is a schematic diagram showing the configuration of the presentation system 1D in this modified example.

As shown in FIG. 23, the presentation system 1D includes a server 10D, a dryer 20A, and a terminal 30D.

Server 10D is a server used in place of server 10 in embodiment 1. Server 10D includes a communication unit 11 and a training unit 12. The communication unit 11 and training unit 12 included in server 10D are the same as those included in server 10 in embodiment 1.

Dryer 20A is similar to dryer 20A in embodiment 1.

Terminal 30D is a terminal used in place of terminal 30A in embodiment 1. Terminal 30D includes a communication unit 31, a control unit 32, a display unit 33, a prediction unit 34, a reliability calculation unit 35, a threshold determination unit 36, and a display control unit 37.

The communication unit 31, the control unit 32, and the display unit 33 are the same as those provided in the terminal 30A in embodiment 1.

In addition, the prediction unit 34, the reliability calculation unit 35, the threshold determination unit 36, and the display control unit 37 are the same as the prediction unit 13, the reliability calculation unit 14, the threshold determination unit 15, and the display control unit 16 provided in the server 10 in the embodiment, respectively.

Note that there may be any number of dryers or terminals. Also, the number of dryers and terminals may differ.

The presentation system 1D in this modified example can be said to be a system in which the functions of the prediction unit 13, reliability calculation unit 14, threshold determination unit 15, and display control unit 16 of the presentation system 1 in the embodiment have been moved from the server to the terminal.

In this way, the calculation of predicted values using the prediction model, the calculation of reliability and thresholds, and display control are performed by terminal 30D, which has the effect of reducing the processing load on server 10D.

(Second Modification of the Embodiment)
In this modification, a presentation method and system for appropriately presenting the timing when a device ends a process, which is different from the above embodiment, will be described.

Figure 24 is a schematic diagram showing the configuration of the presentation system 1E in this modified example.

As shown in FIG. 24, the presentation system 1E includes a dryer 20A and a terminal 30D. The presentation system 1E does not include a server.

Dryer 20A is similar to dryer 20A in embodiment 1.

Terminal 30E is a terminal used in place of terminal 30A in embodiment 1. Terminal 30E includes a communication unit 31, a control unit 32, a display unit 33, a prediction unit 34, a reliability calculation unit 35, a threshold determination unit 36, a display control unit 37, and a training unit 38.

The communication unit 31, the control unit 32, and the display unit 33 are the same as those provided in the terminal 30A in embodiment 1.

In addition, the prediction unit 34, the reliability calculation unit 35, the threshold determination unit 36, the display control unit 37, and the training unit 38 are the same as the prediction unit 13, the reliability calculation unit 14, the threshold determination unit 15, the display control unit 16, and the training unit 12 provided in the server 10 in the embodiment, respectively.

Note that there may be any number of dryers or terminals. Also, the number of dryers and terminals may differ.

The presentation system 1E in this modified example can be said to be a system in which the functions of the training unit 12, prediction unit 13, reliability calculation unit 14, threshold determination unit 15, and display control unit 16 of the presentation system 1 in the embodiment have been moved from the server to the terminal.

In this way, the terminal 30D will train the predictive model, calculate the predicted values using the predictive model, calculate the reliability and thresholds, and control the display, which has the effect of eliminating the need for the server 10.

As described above, the presentation method of the above embodiment or the above modified example presents a predicted value of the timing when the device will end processing from the parameter values of the device using a machine learning model, and notifies the user if the reliability of the predicted value is relatively low. The reliability indicates the degree to which the predicted value is valid, in other words, the degree to which the predicted value is reliable. The user visually recognizes the presented predicted value of the end timing, and if a notification is given, the user can understand that the reliability of the predicted value is relatively low. In this way, the presentation method can appropriately present the timing when the device will end processing.

In addition, the presentation method uses the condition that the first value or the second value is equal to or less than a predetermined value to more easily determine a threshold value used to determine the reliability of the predicted value of the end timing. Therefore, the presentation method can appropriately present the timing at which the device will end processing by easily determining the reliability threshold value.

In addition, the presentation method uses the condition that the first value or the second value is a minimum value to more easily determine the threshold value used to determine the reliability of the predicted value of the end timing. Therefore, the presentation method can appropriately present the timing at which the device will end processing by easily determining the reliability threshold value.

The above presentation method also notifies the user by changing the display mode of the predicted value displayed on the display screen. By visually checking the change in the display mode of the predicted value, the user can easily understand that the reliability of the predicted value is relatively low. Thus, the above presentation method can appropriately present the timing for the device to end processing by notifying the user of the reduced reliability by changing the display mode.

Furthermore, the above presentation method adjusts the threshold value used to determine the reliability of the predicted value of the end timing based on instructions received from the user, so that notifications can be provided at an amount or frequency according to the instructions. The user can receive notifications at an amount or frequency according to the instructions. Thus, the above presentation method can appropriately present the timing at which the device will end processing using notifications of an appropriate amount or frequency.

Furthermore, the above presentation method sets the threshold value used to determine the reliability of the predicted value of the end timing based on instructions received from the user, so that notifications can be provided at an amount or frequency according to the instructions. The user can receive notifications at an amount or frequency according to the instructions. Thus, the above presentation method can appropriately present the timing at which the device will end processing using notifications of an appropriate amount or frequency.

In addition, the above presentation method easily obtains the reliability according to the accuracy of the sensor that obtains the parameter value. Therefore, the above presentation method can appropriately present the timing when the device will end the process.

In addition, the above presentation method can appropriately present the timing when the dryer will finish the process of drying the laundry.

As described above, an embodiment has been described as an example of the technology disclosed herein. For this purpose, the accompanying drawings and detailed description have been provided.

Therefore, among the components described in the attached drawings and detailed description, not only are there components essential for solving the problem, but there may also be components that are not essential for solving the problem in order to illustrate the above implementation. Therefore, the fact that those non-essential components are described in the attached drawings or detailed description should not be used to immediately determine that those non-essential components are essential.

Furthermore, since the above-described embodiments are intended to illustrate the technology disclosed herein, various modifications, substitutions, additions, omissions, etc. may be made within the scope of the claims or their equivalents.

The present disclosure is applicable to a system that predicts the completion timing of equipment processing.

1, 1D, 1E Presentation system 10, 10D Server 11, 21, 31 Communication unit 12, 38 Training unit 13, 34 Prediction unit 14, 35 Reliability calculation unit 15, 36 Threshold determination unit 16, 37 Display control unit 20A, 20B, 20C Dryer 22 Processing unit 23, 33 Display unit 30A, 30B, 30C, 30D, 30E Terminal 32 Control unit 51, 51A, 51B, 51C, 51D, 52, 52A, 52B, 52C, 52D, 53, 54, 61, 62, 71, 72, 81, 82 Image 63, 73, 83 Slider 64, 74, 84 Handle N Network

Claims (9)

A method for presenting a timing at which a device will end a process, comprising the steps of:
a parameter value of the device related to a first process, which is the process, is input to a prediction model having a timing when the device ends the first process as an output, and a parameter value related to each of a plurality of second processes, which are the processes, previously executed by the device, is input to the prediction model, which has a parameter value of the device related to a first process, which is the process, and a timing when the device ends the first process as an output, to obtain a first predicted value of the timing when the device ends each of a plurality of second processes, which is output;
Calculating a first reliability of the first predicted value obtained for each of a plurality of times of the second process;
Acquire actual measurement values at the timing when the device finishes each of the second processes multiple times,
determining a threshold value for the first reliability using the first predicted value, the first reliability, and the actual measured value for each of a plurality of times of the second process;
obtaining a second predicted value indicating a timing at which the device will end the third process, the second predicted value being output by inputting a parameter value related to the third process, which is the process, into the prediction model;
Calculating a second confidence level of the second predicted value;
performing a presentation control for presenting the second predicted value and a notification control for notifying the user when the second reliability is less than the threshold;
In determining the threshold value,
determining the threshold value at which at least one of a first condition and a second condition is satisfied;
the first condition is a condition on a first value that is a ratio of one or more second processes in which a difference between the first predicted value and the actual measured value is less than a tolerance value among N second processes in which the first reliability is lower than the threshold value among a plurality of second processes, where N is an integer equal to or greater than 1;
The second condition is a condition on a second value which is the proportion of one or more second processes in which the first reliability is higher than the threshold value among M second processes in which the difference between the first predicted value and the actual measured value among multiple second processes is greater than or equal to the tolerance value, and M is an integer greater than or equal to 1. The first condition includes a condition that the first value is equal to or less than a predetermined value, or
The presentation method according to claim 1 , wherein the second condition includes a condition that the second value is equal to or less than a predetermined value. The first condition includes a condition that the first value is a minimum value among a plurality of first value candidates calculated using a plurality of candidates for the threshold value, or
The presentation method according to claim 1 , wherein the second condition includes a condition that the second value is a minimum value among a plurality of second value candidates calculated using a plurality of candidates for the threshold value. the presentation control includes control of displaying the second predicted value on a display screen;
The presentation method according to claim 1 , wherein the notification control includes control for changing a display mode of the second predicted value. When determining the threshold value,
Furthermore, an instruction to increase or decrease the determined threshold value is received from a user;
increasing or decreasing the determined threshold value in accordance with the received instruction;
In the notification control,
The presentation method according to claim 1 or 2, further comprising: controlling the notification using the increased or decreased threshold value. When determining the threshold value,
Furthermore, an instruction to set the threshold value is received from a user,
Setting the threshold value according to the received instruction;
In the notification control,
The presentation method according to claim 1 or 2, further comprising: controlling the notification using the threshold value after the threshold value has been set. moreover,
When obtaining the first reliability,
Acquire accuracy of a sensor that has acquired a parameter value of the device related to each of the second processes, and acquire the first reliability that is higher as the accuracy of the acquired sensor is higher;
When obtaining the second reliability,
The presentation method according to claim 1 or 2, further comprising acquiring an accuracy of a sensor that has acquired the parameter value of the device related to the third process, and acquiring a higher second reliability as the accuracy of the acquired sensor is higher. The apparatus is a dryer,
The presentation method according to claim 1 or 2, wherein the process is a process of drying laundry by the dryer. A presentation system that presents a timing at which a device ends a process, comprising:
The present invention includes a prediction unit, a reliability calculation unit, a threshold determination unit, and a presentation control unit,
the prediction unit inputs a parameter value of the device related to a first process, which is the process, into a prediction model having a timing at which the device ends the first process as an output, and inputs parameter values related to each of a plurality of second processes, which are the processes, previously executed by the device, to obtain a first predicted value of the timing at which the device ends each of a plurality of second processes, which is output;
the reliability calculation unit calculates a first reliability of the first predicted value obtained for each of a plurality of times of the second process;
the threshold determination unit acquires actual measurement values at the timings when the device finishes each of the multiple executions of the second process, and determines a threshold value related to the first reliability using the first predicted value, the first reliability, and the actual measurement value for each of the multiple executions of the second process;
the prediction unit obtains a second predicted value indicating a timing at which the device will end the third process, the second predicted value being output by inputting a parameter value related to the third process, which is the process, into the prediction model;
The reliability calculation unit calculates a second reliability of the second predicted value,
The presentation control unit performs presentation control to present the second predicted value, and performs notification control to provide a notification when the second reliability is less than the threshold value;
In determining the threshold value,
determining the threshold value at which at least one of a first condition and a second condition is satisfied;
the first condition is a condition on a first value that is a ratio of one or more second processes in which a difference between the first predicted value and the actual measured value is less than a tolerance value among N second processes in which the first reliability is lower than the threshold value among a plurality of second processes, where N is an integer equal to or greater than 1;
The second condition is a condition on a second value which is the proportion of one or more second processes in which the first reliability is higher than the threshold value out of M number of second processes in which the difference between the first predicted value and the actual measured value is greater than or equal to the tolerance, and M is an integer greater than or equal to 1.

Images