Remaining amount prediction system and remaining amount prediction program

Abstract

To provide a remaining amount prediction system and a remaining amount prediction program capable of improving the accuracy of predicting a date on which the amount of remaining consumable goods falls below a predetermined amount.SOLUTION: A remaining amount prediction system includes a remaining amount prediction unit which predicts an empty date on which a remaining amount of toner used in an image forming apparatus falls to 0%. The remaining amount prediction unit is configured to switch a prediction method as a method for predicting the toner empty date based on an actual toner level, from a linear approximation usage (S143) as a method using linear approximation to a machine learning usage (S147) as a method using a machine learning model, during the process of actual toner depletion (Yes in S142, Yes in S146 or S144, and S145).SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a remaining amount prediction system and a remaining amount prediction program for predicting the remaining amount of consumables used in electronic devices. [Background technology]

[0002] BACKGROUND ART Conventionally, there is known a remaining amount prediction system that predicts the remaining amount of a consumable item used in an electronic device using a machine learning model that predicts the remaining amount of the consumable item (see, for example, Patent Document 1).

[0003] For example, in a conventional remaining amount prediction system, when a machine learning model is used to predict the remaining amount of toner one day after the latest date among dates corresponding to one or more days' worth of remaining amounts of consumables that form the basis of explanatory variables input into the machine learning model, when predicting the date on which the remaining amount of a consumable will fall below a specific amount, the remaining amount of the consumable one day after the latest date indicated in the actual measurement value data can be predicted by first inputting explanatory variables into the machine learning model that are generated based on the daily remaining amounts of the consumables indicated in actual measurement value data that indicate actual measurements of the remaining amounts of the consumables in an electronic device once a day. Next, the remaining amount of the consumable two days after the latest date indicated in the actual measurement value data can be predicted by inputting explanatory variables into the machine learning model that are generated based on the daily remaining amounts of the consumables indicated in the actual measurement value data and the already predicted remaining amounts of the consumables one day after the latest date indicated in the actual measurement value data. Next, by inputting explanatory variables generated based on the remaining amount of the consumable for each day shown in the actual measurement data and the previously predicted remaining amounts of the consumable for one day and two days after the most recent date shown in the actual measurement data into a machine learning model, it is possible to predict the remaining amount of the consumable for three days after the most recent date shown in the actual measurement data. By repeating the above-described operations, it is possible to predict the day on which the remaining amount of the consumable will fall below a specific amount. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-151879 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in conventional remaining quantity prediction systems, due to the nature of machine learning models, the more recent the remaining quantity of a consumable item among the remaining quantities of one or more days that form the basis of the explanatory variables, the greater the impact on the prediction of the remaining quantity of the consumable item.Therefore, depending on the content of the remaining quantities of one or more days that form the basis of the explanatory variables, there is a problem that the day when the remaining quantity of the consumable item will fall below a certain amount may be predicted to be a day that is extremely earlier or extremely later than the day when the remaining quantity of the consumable item will actually fall below the certain amount.

[0006] Therefore, an object of the present invention is to provide a remaining amount prediction system and a remaining amount prediction program that can improve the accuracy of predicting the day when the remaining amount of a consumable item will fall below a specific amount. [Means for solving the problem]

[0007] The remaining quantity prediction system of the present invention includes a remaining quantity prediction unit that predicts the day on which the remaining quantity of a consumable used in an electronic device will fall below a specific amount, and the remaining quantity prediction unit switches the prediction method for predicting the day on which the remaining quantity of the consumable will fall below the specific amount based on the actual remaining quantity of the consumable from a linear approximation usage method that uses linear approximation to a machine learning model usage method that uses a machine learning model as the actual remaining quantity of the consumable decreases.

[0008] With this configuration, the remaining quantity prediction system of the present invention switches the prediction method for predicting the day when the remaining quantity of a consumable will fall below a specific amount based on the actual remaining quantity of the consumable from a linear approximation method using a linear approximation to a machine learning model method using a machine learning model as the actual remaining quantity of the consumable decreases.Therefore, when there are relatively many days between the current date and the day when the remaining quantity of the consumable will fall below a specific amount, the possibility of predicting the day when the remaining quantity of the consumable will fall below a specific amount using the machine learning model method can be reduced, and as a result, the accuracy of predicting the day when the remaining quantity of the consumable will fall below a specific amount can be improved.

[0009] In the remaining quantity prediction system of the present invention, the remaining quantity prediction unit may switch the prediction method from the linear approximation method to the machine learning model method when the number of days from the current date until the date on which the remaining quantity of the consumable will be below the specific amount, as predicted using the linear approximation, is below a specific number of days.

[0010] With this configuration, the remaining quantity prediction system of the present invention switches the prediction method from the linear approximation method to the machine learning model method when the number of days from the current date until the day when the remaining quantity of the consumable will be below a specific amount, as predicted using linear approximation, is below a specific number of days.Therefore, when there are relatively many days from the current date until the day when the remaining quantity of the consumable will be below a specific amount, the possibility of predicting the day when the remaining quantity of the consumable will be below a specific amount using the machine learning model method can be reduced.

[0011] In the remaining quantity prediction system of the present invention, the remaining quantity prediction unit may switch the prediction method from the linear approximation usage method to the machine learning model usage method when the actual remaining quantity of the consumable is below a specific remaining quantity.

[0012] With this configuration, the remaining amount prediction system of the present invention switches the prediction method from the linear approximation usage method to the machine learning model usage method when the actual remaining amount of the consumable is below a specific remaining amount.Therefore, when the linear approximation usage method becomes inappropriate, for example, because the actual remaining amount of the consumable is so low that it can no longer be detected properly, the possibility of predicting the day when the remaining amount of the consumable will be below a specific amount using the machine learning model usage method can be improved.

[0013] The remaining quantity prediction program of the present invention causes a computer to implement a remaining quantity prediction unit that predicts the day on which the remaining quantity of a consumable used in an electronic device will fall below a specific amount, and the remaining quantity prediction unit is characterized in that it switches the prediction method for predicting the day on which the remaining quantity of the consumable will fall below the specific amount based on the actual remaining quantity of the consumable from a linear approximation usage method that uses linear approximation to a machine learning model usage method that uses a machine learning model as the actual remaining quantity of the consumable decreases.

[0014] With this configuration, a computer executing the remaining quantity prediction program of the present invention switches the prediction method for predicting the day on which the remaining quantity of a consumable will fall below a specific amount based on the actual remaining quantity of the consumable from a linear approximation method using linear approximation to a machine learning model method using a machine learning model as the actual remaining quantity of the consumable decreases.This reduces the possibility of using a machine learning model to predict the day on which the remaining quantity of the consumable will fall below a specific amount when there are relatively many days between the current date and the day on which the remaining quantity of the consumable will fall below a specific amount, thereby improving the accuracy of predicting the day on which the remaining quantity of the consumable will fall below a specific amount. [Effects of the Invention]

[0015] The remaining amount prediction system and remaining amount prediction program of the present invention can improve the accuracy of predicting the day when the remaining amount of a consumable item will fall below a specific amount. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a block diagram of a system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram of the remaining amount prediction system shown in FIG. 1 when configured by one computer. [Figure 3] 3 is a diagram showing an example of the remaining amount of toner for each day indicated in the actual measurement value data shown in FIG. 2. FIG. [Figure 4] FIG. 2 is a block diagram of the image forming apparatus shown in FIG. 1 when it is an MFP. [Figure 5] 2 is a sequence diagram of the operation of the system shown in FIG. 1 when a new toner container is set in the image forming apparatus. [Figure 6] 2 is a sequence diagram of the operation of the system shown in FIG. 1 when the remaining amount prediction system records the actual measured value of the remaining amount of toner in the image forming apparatus. [Figure 7] 3 is a flowchart illustrating an example of the operation of the remaining amount prediction system shown in FIG. 2 when predicting the toner empty date of the image forming apparatus. [Figure 8] FIG. 8 is an explanatory diagram of a method for predicting an empty day using linear approximation in the operation shown in FIG. 7. [Figure 9] 8 is a flowchart of the machine learning model use prediction process shown in FIG. 7. [Figure 10] 8 is a flowchart illustrating an example of the operation of the remaining amount prediction system when predicting the toner empty date of the image forming apparatus, which is different from the example shown in FIG. 7. DETAILED DESCRIPTION OF THE INVENTION

[0017] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0018] First, the configuration of a system according to an embodiment of the present invention will be described.

[0019] FIG. 1 is a block diagram of a system 10 according to the present embodiment.

[0020] As shown in Fig. 1, system 10 includes a remaining amount prediction system 20 that predicts the remaining amount of toner, which is a consumable item used in an image forming apparatus, which is an electronic device. Remaining amount prediction system 20 may be configured with a single computer such as a PC (Personal Computer), or may be configured with multiple computers. Remaining amount prediction system 20 may also be configured on the cloud.

[0021] The system 10 includes an image forming device 30, such as an MFP (Multifunction Peripheral) or a dedicated printer. The system 10 can include at least one other image forming device having a similar configuration to the image forming device 30.

[0022] The remaining amount prediction system 20 and the image forming apparatus can communicate with each other via a network 11 such as a LAN (Local Area Network) or the Internet.

[0023] FIG. 2 is a block diagram of a remaining amount prediction system 20 configured by one computer.

[0024] As shown in Figure 2, the remaining amount prediction system 20 includes an operation unit 21, which is an operation device such as a keyboard or mouse through which various operations are input; a display unit 22, which is a display device such as an LCD (Liquid Crystal Display) that displays various information; a communication unit 23, which is a communication device that communicates with external devices via a network such as a LAN or the Internet, or directly via a wired or wireless connection without using a network; a memory unit 24, which is a non-volatile memory device such as a semiconductor memory or an HDD (Hard Disk Drive) that stores various information; and a control unit 25 that controls the entire remaining amount prediction system 20.

[0025] The storage unit 24 can store a remaining amount prediction program 24a for predicting the remaining amount of toner in the image forming apparatus 30. The remaining amount prediction program 24a may be installed in the remaining amount prediction system 20 during the manufacturing stage of the remaining amount prediction system 20, or may be additionally installed in the remaining amount prediction system 20 from an external storage medium such as a USB (Universal Serial Bus) memory, or may be additionally installed in the remaining amount prediction system 20 from a network.

[0026] The storage unit 24 can store a one-day remaining amount prediction model 24b as a machine learning model that predicts the remaining amount of toner one day after the latest base date, which is the latest date among dates corresponding to one or more days' worth of remaining toner amounts that serve as the basis for the explanatory variables input to the machine learning model. GBDT (Gradient Boosting Decision Tree) is used for learning and prediction of the one-day remaining amount prediction model 24b. GBDT is a prediction method that combines (ensembles) multiple "decision trees." The explanatory variables of the one-day remaining amount prediction model 24b are, for example, nine types: the remaining toner amount on the latest reference date, the remaining toner amount one day before the latest reference date, the remaining toner amount two days before the latest reference date, the remaining toner amount three days before the latest reference date, the remaining toner amount four days before the latest reference date, the remaining toner amount five days before the latest reference date, the remaining toner amount six days before the latest reference date, the remaining toner amount 13 days before the latest reference date, and the remaining toner amount 89 days before the latest reference date. The remaining toner amount on days earlier than the latest reference date has little effect on the prediction of the remaining toner amount one day after the latest reference date, so it may be omitted from the explanatory variables to reduce the processing load for predicting the remaining toner amount one day after the latest reference date. The one-day remaining amount prediction model 24b can predict the remaining amount of toner one day after the base latest date if at least one of the nine explanatory variables mentioned above, including the remaining amount of toner on the base latest date, is input.

[0027] The storage unit 24 can store actual measurement value data 24c indicating the actual daily measurement value of the remaining amount of toner in the image forming device 30. Similarly, the storage unit 24 can store actual measurement value data indicating the actual daily measurement value of the remaining amount of toner in the image forming device for each image forming device.

[0028] FIG. 3 is a diagram showing an example of the remaining amount of toner for each day indicated in the actual measurement data 24c.

[0029] As shown in Fig. 3, the actual measurement data 24c shows the date and the actual measurement value of the remaining amount of toner for each day. In the example shown in Fig. 3, the actual measurement values ​​of the remaining amount of toner for eight days from January 1 to January 8 are shown. The remaining amount of toner is expressed as a percentage indicated by an integer between 100 and 0.

[0030] 2 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores programs and various data, and a RAM (Random Access Memory) that serves as a memory used as a work area for the CPU of the control unit 25. The CPU of the control unit 25 executes programs stored in the storage unit 24 or the ROM of the control unit 25.

[0031] The control unit 25 executes the remaining amount prediction program 24a to realize a remaining amount prediction unit 25a that predicts the remaining amount of toner using the one-day remaining amount prediction model 24b.

[0032] FIG. 4 is a block diagram of the image forming apparatus 30 in the case of an MFP.

[0033] As shown in FIG. 4, the image forming apparatus 30 includes an operation unit 31, which is an operation device such as a button through which various operations are input; a display unit 32, which is a display device such as an LCD that displays various information; a printer 33, which is a printing device that prints images on a recording medium such as paper; a remaining amount sensor 34 that detects the amount of toner remaining in a toner container that stores toner consumed by the printer 33; a scanner 35, which is a reading device that reads images from a document; a communication unit 36, which is a communication device that communicates with an external device via a network such as a LAN or the Internet, or directly via a wired or wireless connection without using a network; a memory unit 37, which is a non-volatile memory device such as a semiconductor memory or HDD that stores various information; and a control unit 38 that controls the entire image forming apparatus 30.

[0034] The printer 33 has a replaceable toner container.

[0035] The remaining amount of toner detected by the remaining amount sensor 34 becomes 100% when a used toner container is replaced with a new toner container in the image forming device 30. The remaining amount sensor 34 can detect the remaining amount of toner only when the remaining amount of toner is 5% or more. In other words, the remaining amount sensor 34 cannot detect the remaining amount of toner when the remaining amount of toner is less than 5%.

[0036] Control unit 38 includes, for example, a CPU, a ROM that stores programs and various data, and a RAM as memory used as a work area for the CPU of control unit 38. The CPU of control unit 38 executes programs stored in storage unit 37 or the ROM of control unit 38.

[0037] The operation of the system 10 will now be described.

[0038] First, the operation of the system 10 when a new toner container is set in the image forming apparatus 30 will be described.

[0039] FIG. 5 is a sequence diagram of the operation of the system 10 when a new toner container is set in the image forming apparatus 30. As shown in FIG.

[0040] As shown in FIG. 5, when a new toner container is set in the image forming apparatus 30, the image forming apparatus 30 notifies the remaining amount prediction system 20 that a new toner container has been set (S101).

[0041] When the remaining amount predicting unit 25a of the remaining amount predicting system 20 is notified by the image forming device 30 that a new toner container has been set, the remaining amount predicting unit 25a erases all values ​​in the actual measurement value data 24c (S102).

[0042] When the process of S102 is completed, the remaining amount predicting unit 25a turns off the machine learning model use flag, which is a flag indicating that the remaining amount of toner is predicted using a machine learning model (S103).

[0043] Next, the operation of the remaining amount prediction system 20 when the remaining amount prediction system 20 records the actual measured value of the remaining amount of toner in the image forming device 30 will be described.

[0044] FIG. 6 is a sequence diagram of the operation of the remaining amount prediction system 20 when the remaining amount prediction system 20 records the actual measured value of the remaining amount of toner in the image forming device 30.

[0045] As shown in FIG. 6, the image forming device 30 transmits the actual measured value of the remaining amount of toner each day to the remaining amount prediction system 20, for example, by transmitting the latest actual measured value of the remaining amount of toner detected by the remaining amount sensor 34 before the end of the previous day after the end of each day to the remaining amount prediction system 20 (S121).

[0046] When the actual measurement value of the remaining amount of toner is transmitted from the image forming device 30, the remaining amount prediction unit 25a of the remaining amount prediction system 20 records the transmitted actual measurement value in the actual measurement value data 24c of the image forming device 30 (S122).

[0047] The above describes the operation of the remaining amount prediction system 20 when recording the actual measured value of the remaining amount of toner in the image forming device 30, but the operation of the remaining amount prediction system 20 when recording the actual measured value of the remaining amount of toner in an image forming device other than the image forming device 30 is also similar.

[0048] Next, the operation of the remaining amount prediction system 20 when predicting an empty day when the remaining amount of toner in the image forming device 30 is 0%, that is, when the image forming device 30 is empty, will be described.

[0049] FIG. 7 is a flowchart showing an example of the operation of the remaining amount prediction system 20 when predicting the toner empty date of the image forming device 30.

[0050] As shown in FIG. 7, the remaining amount predicting unit 25a determines whether or not the machine learning model use flag is set (S141).

[0051] When the remaining amount prediction unit 25a determines in S141 that the machine learning model use flag is not set, it determines whether the remaining toner amount α for the latest date among the remaining toner amounts for each day shown in the actual measurement value data 24c is less than 10% of the specific remaining amount (S142).

[0052] When the remaining amount prediction unit 25a determines in S142 that the remaining toner amount α for the latest date among the remaining toner amounts for each day shown in the actual measurement value data 24c is not 10% or less, i.e., is greater than 10%, it predicts the empty day using linear approximation (S143).

[0053] FIG. 8 is an explanatory diagram of a method for predicting empty days using linear approximation in the operation shown in FIG.

[0054] 8, remaining amount prediction unit 25a calculates the slope of a graph with the date on the horizontal axis and the remaining toner amount on the vertical axis based on the remaining toner amounts for 90 days from 90 days before the current date to one day before the current date among the remaining toner amounts for each day shown in actual measurement value data 24c, draws an approximate line of the calculated slope so that it passes through the remaining toner amount one day before the current date, and determines the date on this approximate line where the remaining toner amount is predicted to be 0% or less as the empty date. Note that if the actual measurement value data 24c contains only remaining toner amounts for less than 90 days among the remaining toner amounts for 90 days from 90 days before the current date to one day before the current date, remaining amount prediction unit 25a draws an approximate line based on the remaining toner amount for less than 90 days among the remaining toner amounts for 90 days from 90 days before the current date to one day before the current date.

[0055] As shown in FIG. 7, when the process of S143 is completed, the remaining amount predicting unit 25a determines whether the number of days β from the current date to the date predicted in S143 is 40 days or less (S144).

[0056] If the remaining amount predicting unit 25a determines in S144 that the number of days β from the current date to the date predicted in S143 is 40 days or less, it sets a machine learning model use flag (S145).

[0057] When remaining amount predicting unit 25a determines in S142 that the remaining amount α of toner for the latest date among the remaining amounts of toner for each day indicated in actual measurement value data 24c is 10% or less, remaining amount predicting unit 25a sets a machine learning model use flag (S146).

[0058] When the remaining amount prediction unit 25a determines in S141 that the machine learning model use flag is set or when the processing of S146 is completed, it executes a machine learning model use prediction process to predict the remaining amount of toner using a machine learning model (S147).

[0059] FIG. 9 is a flowchart of the machine learning model use prediction process shown in FIG.

[0060] As shown in FIG. 9, the remaining amount predicting unit 25a generates explanatory variables for the one-day remaining amount predicting model 24b based on the remaining amount of toner for each day indicated in the actual measurement value data 24c (S161).

[0061] When the processing of S161 is completed, the remaining amount prediction unit 25a inputs the explanatory variables generated just before into the one-day remaining amount prediction model 24b, thereby predicting the remaining amount of toner one day after the latest base date of the explanatory variables generated just before (S162).

[0062] When the process of S162 is completed, the remaining amount predicting unit 25a determines whether the remaining amount of toner predicted immediately before is 0% or less (S163).

[0063] If the remaining amount prediction unit 25a determines in S163 that the remaining amount of toner predicted immediately before is not 0% or less, it generates explanatory variables for the one-day remaining amount prediction model 24b based on the daily remaining amount of toner shown in the actual measurement data 24c and the daily remaining amount of toner predicted in all S162 in the current machine learning model usage prediction process shown in Figure 9 (S164), and executes the process of S162.

[0064] If the remaining amount prediction unit 25a determines in S163 that the most recently predicted remaining amount of toner is 0% or less, it determines the date one day after the most recent base date of the explanatory variable generated immediately before as the toner empty date (S165) and terminates the machine learning model usage prediction process shown in Figure 9.

[0065] As shown in FIG. 7, the remaining amount prediction unit 25a terminates the operation shown in FIG. 7 when it determines in S144 that the number of days β from the current date to the date predicted in S143 is not 40 days or less, i.e., is more than 40 days, or when the processing of S145 or S147 is completed.

[0066] As described above, the remaining amount prediction system 20 switches the prediction method for predicting the empty date based on the actual remaining amount of toner from the linear approximation method (S143), which uses linear approximation, to the machine learning model method (S147), which uses a machine learning model, as the actual remaining amount of toner decreases (YES in S142 and S146, or YES in S144 and S145).Therefore, when there are relatively many days between the current date and the empty date, the possibility of predicting the empty date using the machine learning model method can be reduced, and as a result, the accuracy of the empty date prediction can be improved.

[0067] The remaining amount prediction system 20 can improve the likelihood of predicting the empty date using a linear approximation method, which takes less time to process predictions than a machine learning model method, when the number of days from the current date to the empty date is relatively large, thereby shortening the time required to process predictions.

[0068] The remaining amount prediction system 20 switches the prediction method from the linear approximation method to the machine learning model method (S145) when the number of days β from the current date until the empty date predicted using linear approximation is 40 days or less (YES in S144), thereby reducing the possibility of predicting the empty date using the machine learning model method when the number of days from the current date until the empty date is relatively large.

[0069] When the actual remaining amount α of toner is 10% or less (YES in S142), the remaining amount prediction system 20 switches the prediction method from the linear approximation method to the machine learning model method (S146). Therefore, when the actual remaining amount of toner becomes so low that it can no longer be properly detected by the remaining amount sensor 34, the linear approximation method becomes inappropriate, thereby improving the likelihood of predicting the empty date using the machine learning model method.

[0070] In this embodiment, when the actual remaining toner amount α is 10% or less (YES in S142), the remaining amount prediction system 20 switches the prediction method from the linear approximation method to the machine learning model method (S146). The 10% used as the judgment standard in S142 is set as a value equal to or greater than 5% because the lower limit of detection by the remaining amount sensor 34 is 5%. The judgment standard in S142 may be a value other than 10%, as long as it is equal to or greater than the lower limit of detection by the remaining amount sensor 34.

[0071] In this embodiment, the remaining amount prediction system 20 switches the prediction method from the linear approximation method to the machine learning model method based on the remaining toner amount α for the most recent date among the remaining toner amounts for each day indicated in the actual measurement value data 24c (YES in S142 and S146). However, the remaining amount prediction system 20 does not have to switch the prediction method from the linear approximation method to the machine learning model method based on the remaining toner amount α for the most recent date among the remaining toner amounts for each day indicated in the actual measurement value data 24c. For example, the remaining amount prediction system 20 may perform the operation shown in FIG. 10 instead of the operation shown in FIG. 7.

[0072] FIG. 10 is a flowchart of an example, different from the example shown in FIG. 7, of the operation of the remaining amount prediction system 20 when predicting the toner empty date of the image forming device 30.

[0073] As shown in FIG. 10, the remaining amount predicting unit 25a executes the process of S181, which is the same as the process of S141 (see FIG. 7).

[0074] If the remaining amount predicting unit 25a determines in S181 that the machine learning model use flag is not set, it executes the processes of S182 to S184, which are similar to the processes of S143 to S145 (see FIG. 7), and ends the operation shown in FIG. 10.

[0075] If the remaining amount predicting unit 25a determines in S181 that the machine learning model use flag is set, it executes the process of S185, which is the same as the process of S147 (see FIG. 7), and ends the operation shown in FIG.

[0076] In this embodiment, when the number of days β from the current date to the empty date predicted using linear approximation is 40 days or less, the remaining amount prediction system 20 switches the prediction method from the linear approximation method to the machine learning model method (YES in S144 and S145, or YES in S183 and S184). However, the criterion for the determination in S144 or S183 may be a value other than 40 days.

[0077] In this embodiment, the remaining amount prediction system 20 calculates the slope of the approximate line based on the remaining toner amounts for 90 days from 90 days before the current date to 1 day before the current date, among the daily remaining toner amounts indicated in the actual measurement value data 24c. However, the remaining amount prediction system 20 may also calculate the slope of the approximate line based on the remaining toner amounts for 45 days from 45 days before the current date to 1 day before the current date, among the daily remaining toner amounts indicated in the actual measurement value data 24c.

[0078] In this embodiment, remaining amount prediction system 20 generates explanatory variables of the machine learning model based on nine types of remaining toner amounts among the daily remaining toner amounts indicated in actual measurement value data 24c: the remaining toner amount on the base latest date, the remaining toner amount one day before the base latest date, the remaining toner amount two days before the base latest date, the remaining toner amount three days before the base latest date, the remaining toner amount four days before the base latest date, the remaining toner amount five days before the base latest date, the remaining toner amount six days before the base latest date, the remaining toner amount 13 days before the base latest date, and the remaining toner amount 89 days before the base latest date (S161 and S164). However, remaining amount prediction system 20 may also generate explanatory variables of the machine learning model based on remaining toner amounts other than the nine types of remaining toner amounts among the daily remaining toner amounts indicated in actual measurement value data 24c.

[0079] In this embodiment, the remaining amount prediction system 20 predicts the empty date. However, the remaining amount prediction system 20 may predict the date when the remaining amount of the consumable item will be equal to or less than a specific amount other than 0%.

[0080] In this embodiment, the consumable of the present invention is toner used in an image forming apparatus, but the consumable of the present invention may be a consumable other than toner used in an image forming apparatus, or may be a consumable used in an electronic device other than an image forming apparatus. [Explanation of symbols]

[0081] 20 Remaining amount prediction system (computer) 24a Remaining amount prediction program 24b 1-day remaining amount prediction model (machine learning model) 25a Remaining amount prediction section 30 Image forming equipment (electronic equipment)

Claims

1. a remaining amount predicting unit that predicts the day when the remaining amount of a consumable used in the electronic device will be equal to or less than a specific amount; The remaining quantity prediction system is characterized in that the remaining quantity prediction unit switches the prediction method for predicting the day when the remaining quantity of the consumable will be below the specified amount based on the actual remaining quantity of the consumable from a linear approximation method using a linear approximation to a machine learning model method using a machine learning model as the actual remaining quantity of the consumable decreases.

2. The remaining quantity prediction system described in claim 1, characterized in that the remaining quantity prediction unit switches the prediction method from the linear approximation usage method to the machine learning model usage method when the number of days from the current date until the date on which the remaining quantity of the consumable will be below the specific amount, predicted using the linear approximation, is less than a specific number of days.

3. The remaining quantity prediction system according to claim 2, characterized in that the remaining quantity prediction unit switches the prediction method from the linear approximation usage method to the machine learning model usage method when the actual remaining quantity of the consumable is below a specific remaining quantity.

4. A remaining amount prediction unit is implemented in a computer to predict the day when the remaining amount of a consumable used in an electronic device will be equal to or less than a specific amount; A remaining quantity prediction program characterized in that the remaining quantity prediction unit switches a prediction method for predicting the day on which the remaining quantity of the consumable will be below the specific amount based on the actual remaining quantity of the consumable from a linear approximation method using a linear approximation to a machine learning model method using a machine learning model as the actual remaining quantity of the consumable decreases.

Images