JP2017049759A - Failure prediction management device and failure prediction management program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress over quality of failure prediction by taking an allowance degree of a failure into account.SOLUTION: For a predetermined period ΔT4 including a discrimination day when the presence of failure occurrence is discriminated, toner consumption data per unit number of sheets are extracted. If the extracted average value is equal to or larger than a threshold that is calculated by a maintenance decision index generation part 34, it is judged that an image forming apparatus 10 subjected to monitoring is mainly outputting photographic images. In a failure prediction notification part 40, a user who manages the image forming apparatus 10 subjected to monitoring is notified of failure occurrence, and maintenance by a service engineer is required. If the average value of the toner consumption data per unit number of sheets for the period ΔT4 is smaller than the threshold value that is calculated by the maintenance decision index generation part 34, it is judged that the image forming apparatus 10 subjected to monitoring is mainly outputting text images. In the failure prediction notification part 40, the user who manages the image forming apparatus 10 subjected to monitoring is not notified of failure occurrence and maintenance by a service engineer is not required, either.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a failure prediction management apparatus and a failure prediction management program.

  For example, in an image forming apparatus that forms and outputs an image on a recording sheet, when a failure (including failure or malfunction) that impedes its operation occurs, inconvenience occurs for the user of the image forming apparatus. Therefore, the occurrence of such a failure is predicted, and by making it possible to perform necessary measures such as parts replacement and repair immediately before the occurrence of the failure or after the failure has occurred, the use of the image forming apparatus can be improved. It would be desirable to reduce the time for a limited state.

  Here, various inventions have been proposed for predicting failure prediction for an apparatus such as an image forming apparatus.

  In Patent Document 1, when a normal operation of the image forming apparatus is performed, an internal state signal obtained by detecting the state of the photosensitive drum of the image forming apparatus is stored in a storage device. Then, a predetermined state discriminant is created from the reflected light state signal of the photosensitive drum. Then, it is described that failure prediction is performed based on whether or not the state determination signal exceeds a predetermined threshold, and whether or not maintenance is performed during the normal operation of the image forming apparatus is determined based on the prediction result. Has been.

  Patent Document 2 stores the distribution of the occurrence frequency of the feature amount when a failure occurs in the image forming apparatus and the distribution of the occurrence frequency of the feature amount when no failure occurs, from the latest machine information. Calculates the value of each feature value, and calculates the failure predictor probability in the monitored image forming apparatus based on the calculated feature value value and each distribution stored in the failure determination probability distribution storage unit In addition, it is described that the predictive probability of occurrence of a failure is notified.

JP 2007-298875 A JP 2013-109483 A

  Failure tolerance for each device is not taken into account, and failure prediction sometimes becomes over quality.

  An object of the present invention is to obtain a failure prediction management device and a failure prediction management program that can suppress the excessive quality of failure prediction by considering the above facts and considering the tolerance of failure.

According to the first aspect of the present invention, the occurrence of a failure is predicted based on the failure occurrence probability calculated by classifying the information related to the maintenance history and the processing history of the monitoring target device into normal time and abnormal time. A determination unit that determines a prediction index and a determination index that serves as an index for determining whether or not the maintenance corresponding to the occurrence of the failure is necessary based on output information related to the image output by the monitoring target device;
And determining means for determining whether maintenance is necessary or not for an apparatus in which the occurrence of a failure is predicted by the predicting means based on the determination index determined by the determining means.

  The invention according to claim 2 is a notification for notifying the occurrence of a failure to the management device that manages the device to be monitored when the occurrence of a failure is predicted by the prediction means in the invention according to claim 1. It is possible to select whether to further notify and always notify when the occurrence of a failure is predicted or to depend on the determination result of the determination unit.

  According to a third aspect of the present invention, in the first or second aspect of the present invention, the prediction means includes a first distribution indicating a distribution of the occurrence frequency of the feature quantity classified at the time of the abnormality, and the normal time. A second distribution indicating the distribution of the frequency of occurrence of the classified feature quantity is stored, and the current value of each feature quantity in the monitoring target device is acquired and collated with the first distribution and the second distribution. To calculate the failure probability.

  According to a fourth aspect of the present invention, in the invention according to any one of the first to third aspects, the output information is at least a toner consumption amount per unit number of sheets, an image formation area, or image density information. A threshold value, which is a piece of numerical information, is set for the numerical information and is an acceptable level of image quality required for the device to be monitored, and the threshold value is used as the determination index.

According to the fifth aspect of the present invention, the information related to the maintenance history and the processing history of the monitoring target device is classified into a normal time and an abnormal time on the computer, and the distribution of the occurrence frequency of the feature amount classified at the abnormal time is shown. Storing a first distribution and a second distribution indicating a distribution of occurrence frequency of the feature amount classified in the normal state, obtaining a value of each current feature amount in the monitoring target device, and obtaining the first distribution and The failure occurrence probability is calculated by collating with the second distribution, and based on the output information regarding the image output by the monitoring target device, a determination index serving as an index for determining the necessity of maintenance corresponding to the failure occurrence is determined.
It is a failure prediction management program for executing, on the basis of the determined determination index, determining whether or not maintenance is required for a device that is predicted to generate a failure.

  According to the first aspect of the present invention, it is possible to suppress over-prediction of failure prediction by considering the tolerance of failure.

  According to the second aspect of the present invention, it is possible to select whether or not to notify the prediction of the occurrence of a failure regardless of whether maintenance is necessary.

  According to the third aspect of the present invention, it is possible to calculate the failure occurrence probability with higher accuracy than in the case where the feature amount is not used.

  According to the fourth aspect of the present invention, a threshold value can be set for each device to be monitored.

  According to the fifth aspect of the present invention, it is possible to suppress the excessive quality of the failure prediction by taking into account the tolerance of the failure.

It is the schematic of the failure prediction system which concerns on this Embodiment. It is a timing chart which shows the data collection time when generating the failure sign determination model which concerns on this Embodiment. It is a front view of a recording sheet, (A) a text image, (B) shows a photographic image. FIG. 4 is a characteristic diagram of toner consumption per unit number of sheets according to the present embodiment-output number characteristics, in which (A) is a distribution diagram of text images, (B) is a distribution diagram of photographic images, and (C) is a threshold of separation. It is a distribution map for obtaining a value. 6 is a control flowchart for generating a failure determination probability distribution in a failure sign determination model generation unit. (A) is a figure which shows the example of frequency distribution of the calculation value of the feature-value in the period when a failure generate | occur | produced, (B) is an example of frequency distribution of the calculated value of the feature-value in the period when a failure did not occur FIG. It is a control flowchart for calculating the failure occurrence probability by the failure occurrence prediction unit (mode 1 to mode 3). It is a control flowchart for calculating the failure occurrence probability by the failure occurrence prediction unit (mode 4). It is a figure which shows the example of notification of generation | occurrence | production predictor probability. It is a control flowchart which shows the maintenance determination parameter | index production | generation process in the maintenance determination parameter | index production | generation part of the alerting | reporting part which concerns on this Embodiment. It is a control flowchart which shows the maintenance object determination process in the maintenance object determination part of the alerting | reporting part which concerns on this Embodiment.

  FIG. 1 shows a configuration example of an environment for executing failure prediction management control according to the present embodiment.

  In the present embodiment, an image forming apparatus 10 that forms and outputs an image on recording paper P (for example, see FIG. 3), and a maintenance information input terminal that is used by an administrator or maintenance staff of the image forming apparatus 10 12. In FIG. 1, two image forming apparatuses 10 and two maintenance information input terminals 12 are shown, but the number of these is arbitrary.

  In this embodiment, each image forming apparatus 10 and maintenance information input terminal 12 are connected to each of the image forming apparatus 10 and the maintenance information input terminal 12 via a communication network or the like so as to be able to communicate with each other by wire or wirelessly. 10 includes a management device 14 as an example of a failure prediction management device that calculates the probability that a failure will occur. In FIG. 1, the management device 14 is configured by a single device, but each functional unit described below may be distributed to a plurality of devices.

  First, the image forming apparatus 10 will be described.

  The image forming apparatus 10 is an apparatus having an image forming function for forming and outputting an image on the recording paper P. Examples of the image forming apparatus 10 include apparatuses such as a printer (document printing apparatus), a copier (document copying apparatus), and a facsimile apparatus (document transfer apparatus). Is also included.

  Here, the image forming apparatus 10 according to the present embodiment has a function of detecting values of a plurality of monitoring parameters indicating the internal state at any time during the image forming operation. The monitoring parameter is predetermined as a parameter that can contribute to the prediction of the occurrence of a failure, and examples thereof include a charging voltage, a developing bias, a laser light quantity, and a toner density. As a detection value of a monitoring parameter, a measurement value measured at a part corresponding to the monitoring parameter may be used, a target value for controlling each part may be used, and a difference between the measurement value and the target value Other types of values may be used.

  In the image forming apparatus 10 according to the present embodiment, when a job command instructing execution of image forming processing for one page or a plurality of pages is received, an image is formed on the recording paper P according to the job command and output (1 The value of each monitoring parameter is detected for each page), and machine information storing the detected value of each monitoring parameter is transmitted to the management device 60 after completion of all the image forming processes related to the job command.

  In this embodiment, the machine information transmitted to the management apparatus 14 for each job command of image forming processing includes a device ID for identifying the own device, a job ID for identifying the job command, and an image forming process based on the job command. The data has a structure storing the detection value and the detection date and time of each monitoring parameter.

  Here, as described above, instead of the configuration in which the machine information is transmitted to the management device 14 at the end of the image forming process based on the job command, the machine information is temporarily stored in the memory, and is determined in advance. A configuration may be adopted in which the accumulated untransmitted machine information is transmitted when the transmission condition is satisfied. Specifically, for example, the passage of a predetermined time may be used as a transmission condition, and transmission may be performed every time (for example, every hour), or a request from the management device 14 may be used as a transmission condition. May be sent in response to

  Next, the maintenance information input terminal 12 will be described.

  The maintenance information input terminal 12 of this embodiment visits the installation location of the image forming apparatus 10 and receives maintenance information related to the maintenance work from a maintenance person who actually performed the maintenance work or a person who received the report. The input is received and transmitted to the management device 14. In addition, the maintenance information input terminal 12 according to the present embodiment receives information on the prediction result of failure occurrence in the image forming apparatus 10 from the management apparatus 14 and displays it on a display device provided in the maintenance information input terminal 12. .

  The maintenance information transmitted to the management apparatus 14 by the maintenance information input terminal 12 according to the present embodiment indicates a device ID for identifying the image forming apparatus 10 that is the object of the maintenance work, and the date and time when the maintenance work was performed. Data with a structure that stores maintenance date and time, failure type indicating the type of failure removed by the maintenance work, failure date and time indicating the date and time when the failure occurred, failure location indicating the location where the failure occurred, etc. Yes. In other words, the maintenance information can be said to be information on a failure occurrence example.

  Next, the management device 14 will be described.

  The management apparatus 14 according to the present embodiment is an apparatus that predicts the occurrence of a failure in the image forming apparatus 10.

  The management device 14 includes a machine information collection unit 18, a maintenance information collection unit 20, a machine information storage unit 22, and a maintenance information storage unit 24 as the data acquisition unit 16.

  In addition, the management device 14 includes a failure sign determination model generation unit 28 and a failure occurrence prediction unit 30 as the model generation unit 26.

  The data acquisition unit 16 and the model generation unit 26 function as an example of a prediction unit.

  Furthermore, the management device 14 includes a maintenance determination index generation unit 34, a maintenance target determination unit 36, and a failure sign notification unit 40 as the notification unit 32. The maintenance determination index generation unit 34 functions as an example of a determination unit, and the maintenance target determination unit 36 functions as an example of a determination unit. Further, the failure sign notification unit 40 functions as an example of a notification unit.

(Data acquisition unit 16)
The data acquisition unit 16 periodically collects machine information including monitoring parameter data, usage environment information, and usage status information from the monitoring target machine (image forming apparatus 10) by the machine information collection unit 18 and the collected machines. Information is stored in the machine information storage unit 22.

  The machine information includes a device ID for identifying the image forming apparatus 10, a job ID for identifying a received job command in the image forming apparatus 10, detection values and detection values of each monitoring parameter for each image forming process based on the job command. Date and time are also included.

  In the data acquisition unit 16, the maintenance information received from the maintenance information input terminal 12 is collected by the maintenance information collection unit 20, and the collected maintenance information is accumulated in the maintenance information accumulation unit 24.

  The maintenance information includes a device ID for identifying the image forming apparatus 10, a maintenance date and time indicating the date and time when the maintenance work was performed on the image forming apparatus 10, and the type of failure removed from the image forming apparatus 10 by the maintenance work. , A failure date and time indicating the date and time when the failure occurred, and a failure occurrence location and the like indicating the location where the failure occurred are included.

(Model generation unit 26)
The failure sign determination model generation unit 28 in the model generation unit 26 extracts a failure occurrence example from the maintenance information storage unit 24, and monitors the target monitoring parameter for each failure type from the machine in which the failure has occurred (image forming apparatus 10). Data is acquired from the machine information storage unit 22 for a predetermined period ΔT1 shown in FIG.

  For example, the period ΔT1 is preferably about a half year in the first half which is a half period when the entire period (ΔT) of data collection is one year. It may be a relatively short period (one job unit, several job units, one day unit, several days unit). Although details will be described later, in this embodiment, when a failure occurrence is predicted, the data collection period (period ΔT3) for determining whether or not to perform maintenance is used as the data collection period (period ΔT3). The latter half of the year is applied (ΔT1≈ΔT3).

  For example, in the case of an image quality failure case related to density fluctuation, data such as a charging voltage, a developing bias, and a laser light quantity are acquired from the machine information storage unit 22 as monitoring parameters.

  Further, feature quantities such as an average value, a standard deviation, and a correlation value between parameters for a predetermined period ΔT1 are generated for each parameter.

  Further, the failure sign determination model generation unit 28 sets a predetermined period ΔT2 shown in FIG. 2 before the failure occurrence date in a predetermined period (for example, about 1 to 2 weeks), and includes the data for the period ΔT1. 2, the predictive determination model in which the data of the period ΔT2 shown in FIG. 2 is abnormal data, the data of the period other than the period ΔT1 is normal data, and the characteristic amount data is created for normal and abnormal distribution for each monitoring parameter Generate.

  In the failure occurrence prediction unit 30, monitoring parameter data (preferably the latest) of the monitoring target machine (image forming apparatus 10) collected by the machine information collection unit 18 and accumulated in the machine information accumulation unit 22, and a failure sign determination model generation The failure occurrence probability is calculated using the sign determination model generated by the unit 28.

  In other words, based on past failure cases, feature data is divided into an abnormal data group (data group when a failure occurs) and a normal data group (data group when a failure does not occur). The occurrence of a malfunction in the image forming apparatus 10 is predicted using the difference in distribution of various feature value values between the normal time and the normal time.

(Notification part 32)
The notification unit 32 basically notifies the occurrence of a failure based on the failure occurrence predictive probability in the failure prediction target image forming apparatus 10 calculated by the failure occurrence prediction unit 30. Specifically, information on the calculation result (probability of occurrence) is transmitted to the maintenance information input terminal 12 and is displayed and output by a display device provided in the maintenance information input terminal 12.

  In this embodiment, the failure occurrence prediction based on the failure occurrence prediction probability calculated by the failure occurrence prediction unit 30 is adapted to the needs of the user of the machine (image forming apparatus 10). Whether or not notification can be executed is determined after determining whether or not there is.

(An example of conformity to user needs)
In the prediction of the occurrence of a failure in the image forming apparatus 10 based on the prediction in the failure occurrence prediction unit 30, for example, a failure in which an image quality defect such as density unevenness that causes a problem in a graphic image that requires relatively high image quality occurs. Even when the information output by the user using the image forming apparatus 10 is centered on the text, the image quality is not deteriorated, or the allowable range is such that the user sensitivity is not noticed even when the image quality failure occurs. There is.

  In this case, if a service engineer is always visited (maintenance performed) in response to the failure occurrence prediction in the failure occurrence prediction unit 30, it may be perceived as excessive quality and excessive maintenance for the user. In some cases, the function of the image forming apparatus 10 may be temporarily interrupted, leading to a reduction in user productivity.

  Therefore, in the present embodiment, the maintenance determination index generation unit 34 and the maintenance target determination unit 36 in the notification unit 32 determine whether or not the failure predictive probability of the failure prediction target is suitable for the user's needs. ing.

  In the notification unit 32, the monitoring parameter data to be targeted for each type of failure from the image forming apparatus 10 in which the failure has occurred is newly added for a predetermined period ΔT3 shown in FIG. 2 (if ΔT is about one year, the latter half of Of the machine information storage unit 22.

  For the monitoring parameter data acquired from the machine information storage unit 22, a predictive probability is calculated for each day by a predictive determination model.

  The maintenance determination index generation unit 34 extracts toner consumption data per unit number of sheets as an example of image output information from the monitoring parameter data acquired in the period ΔT3. The image output information may be image formation area or image density information, or a combination of two or more thereof, in addition to the toner consumption data per unit number.

  Furthermore, the day when the predictive probability calculated by the failure occurrence prediction unit 30 is equal to or higher than the threshold is extracted.

  In the maintenance determination index generation unit 34, the extracted toner consumption amount data group per number of sheets is included in a predetermined period ΔT2 before the starting point and not included in the period ΔT2 starting from the failure occurrence date. Categorize into things.

  The toner consumption data value per unit number at which the degree of separation between the toner consumption data included in the period ΔT2 and the toner consumption data not included in the period ΔT2 is the maximum value is used as a repair determination index threshold. .

  The degree of separation is calculated by, for example, the following equation (1). Note that means for determining a threshold other than the degree of separation calculation based on the equation (1) may be used as the threshold.

FIG. 3 shows an output form of the recording paper P having different image densities in the toner consumption data group.

  FIG. 3A shows a text-centric output form (hereinafter referred to as “text image”), and the image density is relatively low. FIG. 3B shows an output form centered on a photograph or graphic (hereinafter referred to as “photo image”), and has a relatively high image density.

  FIG. 4A is an output sheet number distribution diagram of the text image shown in FIG. 3A for a specific user.

  FIG. 4B is a distribution diagram of the number of output photographic images shown in FIG. 3B for a specific user.

  In FIGS. 4A and 4B, the horizontal axis represents the toner consumption amount per unit number, but is not limited to the toner consumption amount per unit number, or the printing area per unit number, or It may be the image density per unit number.

  As shown in FIG. 4C, the intersection of the distribution map of the text image and the distribution map of the photographic image is calculated as the degree of separation by the calculation of the equation (1), and this intersection is set as a threshold value. Thus, it can be determined whether the specific user is centered on the text image or the photographic image.

  When the failure occurrence probability calculated by the failure occurrence prediction unit 30 is equal to or greater than a predetermined criterion for failure occurrence, the maintenance target determination unit 36 monitors for a predetermined period ΔT4 (see FIG. 2) including the determination date. Toner consumption data per unit number of the target image forming apparatus 10 is extracted. The period ΔT4 is set to about 1 to 2 weeks as in the case of ΔT2.

  Here, if the extracted average value is equal to or greater than the threshold value calculated by the maintenance determination index generation unit 34, it is determined that the image forming apparatus 10 to be monitored is outputting mainly the photographic image, and the failure is detected. The prediction notification unit 40 notifies the user who manages the image forming apparatus 10 to be monitored of the occurrence of a failure and requests maintenance by a service engineer.

  In other words, if the average value of the toner consumption data per unit number of sheets for the period ΔT4 is less than the threshold value calculated by the maintenance determination index generation unit 34, the monitored image forming apparatus 10 centers on the text image. The failure prediction notification unit 40 does not notify the user who manages the image forming apparatus 10 to be monitored of the occurrence of the failure, and does not request maintenance by a service engineer.

  The operation of this embodiment will be described below.

(Problem determination probability distribution generation process)
Generation of the failure determination probability distribution in the failure sign determination model generation unit 110 will be described with reference to the processing flow illustrated in FIG.

  First, referring to the maintenance information storage unit 108, a model generation target failure occurrence example (maintenance information) is extracted (step 100).

  Next, monitoring parameter data for a period ΔT1 is acquired from the machine information storage unit 22 for each image forming apparatus 10 in which a failure has occurred (step 102).

That is, with reference to the machine information of the machine information storage unit 22 corresponding to the model generation target failure occurrence example (maintenance information), the image forming apparatus 10 in which the failure has occurred (maintenance work has been performed) occurs in that apparatus. the correspondence between failure type is set in advance for each monitoring parameters (and may contribute to the prediction of the occurrence of the type of failure), and acquires the data (detected value) by the period [Delta] T ₁ units. For example, in the case of an image quality failure related to density fluctuation, data such as a charging voltage, a developing bias, and a laser light amount are acquired as monitoring parameters.

Then, for each image forming apparatus 10, it calculates a feature amount based on the data of the monitoring parameter obtained by the period [Delta] T ₁ unit with respect to failure type occurring in the device (step 104).

  The types of feature quantities to be calculated are defined in advance for each fault type, and a plurality of types of feature quantities that change characteristically in relation to the occurrence of the fault of that type are calculated.

  Next, a frequency distribution of each feature quantity is generated in the period ΔT2 up to the date of failure and several days before it and in a period other than that (period ΔT2) (step 106).

  That is, for each of the image forming apparatuses 10, for each of a plurality of types of feature amounts corresponding to the type of failure that has occurred in the device, the frequency of feature value calculation values in the period ΔT2 before the start point, starting from the failure occurrence date and time. A distribution (histogram) and a frequency distribution (histogram) of calculated values of feature amounts in periods other than the period ΔT2 are created.

  For example, the frequency distribution with failure as illustrated in FIG. 6A (frequency distribution of the calculated value of the feature amount in the period when the failure occurred) and the frequency distribution without failure as illustrated in FIG. (Frequency distribution of calculated feature values during a period when no failure occurs) is created. It should be noted that the frequency distribution of the calculated feature value can be created by counting the number of calculated feature values (frequency value) for each section obtained by dividing the range of possible values of the feature value by a certain width.

  For each image forming apparatus 10, frequency values in two frequency distributions (frequency distribution with a failure and frequency distribution without a failure) created for each of a plurality of types of feature amounts corresponding to the types of failures that occurred in the device are obtained. Each result is normalized, and the result is stored (stored) in the failure sign determination model generation unit 28 as a failure type determination probability distribution (step 108).

  In addition, in order to correct the variation in the feature amount between the image forming apparatuses 10, the average value and the standard deviation of each feature amount may be calculated for each image forming apparatus 10, and the frequency distribution may be created by normalizing the feature amount. Good.

  In addition, a failure location determination probability distribution is created by the same procedure (the procedure of the flow of FIG. 5).

  That is, referring to the maintenance information storage unit 24, a failure occurrence example (maintenance information) is extracted.

Next, with reference to the machine information in the machine information storage unit 22 corresponding to the failure occurrence example (maintenance information), the failure occurrence location in the image forming apparatus 10 in which the failure has occurred (maintenance work has been performed) The data (detection value) is acquired for each period ΔT for each monitoring parameter for which the correspondence with ( ₁ ) may be set in advance (which may contribute to the prediction of the occurrence of the failure at that location). For example, in the case of a failure caused by the development system, data such as development bias and toner density are acquired as monitoring parameters.

Then, for each image forming apparatus 10, calculates a feature amount based on the data of the monitoring parameter obtained by the period [Delta] T ₁ unit with respect to point of failure of the device. The types of feature quantities to be calculated are defined in advance for each fault occurrence location, and a plurality of types of feature quantities that change characteristically in relation to the occurrence of the fault at that location are calculated.

Next, for each of the image forming apparatuses 10, for each of a plurality of types of feature amounts corresponding to a failure occurrence location in the apparatus, a calculated value of the feature amount in a period ΔT ₂ before the start point, starting from the failure occurrence date. A frequency distribution (histogram) and a frequency distribution (histogram) of calculated values of feature amounts in periods other than the period ΔT ₂ are created.

  For each image forming apparatus 10, the frequency values in the two frequency distributions (frequency distribution with failure and frequency distribution without failure) created for each of a plurality of types of feature quantities corresponding to the failure occurrence location in the apparatus are respectively shown. The result is normalized, and the result is stored (stored) in the failure sign determination model generation unit 110 as a failure location determination probability distribution.

  In the present embodiment, the normalized frequency distribution for each feature amount corresponding to the failure type is stored in the failure sign determination model generation unit 110 for each image forming apparatus 10 as a failure type determination probability distribution. However, the average frequency distribution is calculated from the normalized frequency distribution for each feature amount corresponding to the failure type for all the image forming apparatuses 10 minutes, and this is used as a failure type determination probability distribution to determine a failure sign You may memorize | store in the model production | generation part 28. FIG.

  In the present embodiment, the normalized frequency distribution for each feature quantity corresponding to the failure occurrence location is stored in the failure sign determination model generation unit 110 for each image forming apparatus 10 as the failure occurrence location determination probability distribution. However, the average frequency distribution is calculated from the normalized frequency distribution for each feature amount corresponding to the failure occurrence locations of all the image forming apparatuses 10 minutes, and this is used as the failure occurrence location determination probability distribution. The failure predictor determination model generation unit 28 may store the failure predictor determination model generation unit 28.

(Failure occurrence probability calculation process)
Next, calculation of the failure occurrence probability by the failure occurrence prediction unit 30 will be described. In the present embodiment, the failure occurrence probability can be calculated by four types [form 1] to [form 4], and an example of a processing flow according to [form 1] to [form 3] is shown in FIG. FIG. 8 shows an example of a processing flow according to [Mode 4].
[Form 1]
As shown in FIG. 7, first, the monitoring parameter data necessary for calculating the feature amount is obtained by referring to the latest machine information stored in the machine information storage unit 22 for the image forming apparatus 10 to be predicted for failure. (Step 110), and the feature amounts are calculated by the same method as the failure sign determination model generation unit 28 (step 112).

  Next, a failure type determination probability distribution is acquired from the failure sign determination model generation unit 28 (step 114), and for each failure type, the probability that that type of failure will occur in the near future of the failure prediction target image forming apparatus 10 is obtained. (Predictive probability of occurrence) is calculated (step 116).

In the present embodiment, the failure type for which the failure occurrence probability is to be calculated is the failure T, and m types of parameters p _j (1 ≦ j ≦ m) included in the latest machine information in the image forming apparatus 10 that is the failure prediction target. Each value of the n types of feature amounts X _i (1 ≦ i ≦ n) related to the failure T calculated based on the failure T is set to x _i , and the failure T The probability of occurrence of is calculated. Note that equation (2) assumes that there is no correlation between the feature quantities.

Here, P (T = yes) is a probability (prior probability) that the failure T occurs, P (T = no) is a probability (prior probability) that the failure T does not occur, and P (T = yes) ) + P (T = no) = 1.

_{Also, P (x i | (T} = yes)) is the probability value of the i-th feature quantity _{X i} when a failure T occurred was _{x i,} the feature quantity _{X i} corresponding to fault T The probability of x _i in the failure type determination probability distribution (with failure) is used.

P (x _i | (T = no)) is a probability that the value of the i-th feature value X _i is x _i when the failure T does not occur, and the feature value corresponding to the failure T using the probability of x _i in the failure type determining a probability distribution over X _i (without failure).

In other words, in the equation (2), the probability (probability) that the failure T occurs and the values of the n types of feature amounts X _i (1 ≦ i ≦ n) when the failure T occurs are expressed as (x ₁ , x ₂ ,..., X _n ) multiplied by the probability that the combination was obtained [P (T = yes) · ΠP (x _i | (T = yes))] and the probability that the failure T does not occur (Priority probability) and a combination of (x ₁ , x ₂ ,..., X _n ) as each value of the n types of feature amounts X _i (1 ≦ i ≦ n) when the failure T does not occur. Using the value [P (T = no) · ΠP (x _i | (T = no))] multiplied by the obtained probability, the probability [P of occurrence of the failure T in the image forming apparatus 10 to be predicted for failure [P ((T = yes) | x ₁ , x ₂ ,..., X _n )] is calculated.

  Thereafter, a preparation process for notifying the occurrence predictor probability calculated for each type of failure is executed (step 118).

  “Preparation for notification” in step 118 indicates that there is a case where notification is not made depending on notification and maintenance determination described later. However, even if a maintenance request is not required, only notification may be executed.

  Note that all the calculated occurrence probabilities may be notified, or only those exceeding a predetermined threshold value may be notified. Further, when notifying a plurality of occurrence predictor probabilities, it is preferable to notify them in descending order of probability. The notification of the occurrence predictor probability is performed by, for example, displaying and outputting with a display device provided in the maintenance information input terminal 12.

FIG. 9A shows an example in which the occurrence predictor probability for each failure type is notified in a list format, and the occurrence predictor probabilities for each failure type are displayed in descending order of probability.
[Form 2]
As shown in FIG. 7, first, the monitoring parameter data necessary for calculating the feature amount is obtained by referring to the latest machine information stored in the machine information storage unit 22 for the image forming apparatus 10 to be predicted for failure. (Step 110), and the feature amounts are calculated by the same method as the failure sign determination model generation unit 28 (step 112).

  Next, a failure location determination probability distribution is acquired from the failure predictor determination model generation unit 28 (step 120), and for each failure occurrence location, a failure at that location will occur in the near future of the image forming apparatus 10 that is the failure prediction target. The probability of occurrence (probability of occurrence) is calculated (step 122).

In the present embodiment, the failure occurrence location for which the failure occurrence probability is to be calculated is the failure T, and n types of feature quantities relating to the failure T calculated based on the latest machine information in the image forming apparatus 10 that is the failure prediction target. Each value of X _i (1 ≦ i ≦ n) is set to x _i , respectively, and the probability of occurrence of a failure T in the failure prediction target image forming apparatus 10 is calculated by equation (2).

  Thereafter, a preparation process for notifying the occurrence predictor probability calculated for each failure occurrence point is executed (step 124).

  “Preparation for notification” in step 124 indicates that notification may not be made depending on notification to be described later and whether maintenance is possible or not, but a maintenance request is not required, but only notification may be executed.

  Note that all the calculated occurrence probabilities may be notified, or only those exceeding a predetermined threshold value may be notified. Further, when notifying a plurality of occurrence predictor probabilities, it is preferable to notify them in descending order of probability. The notification of the occurrence predictor probability is performed by, for example, displaying and outputting with a display device provided in the maintenance information input terminal 12.

FIG. 9B shows an example of notifying the occurrence probabilities for each failure occurrence location in a list format, and the occurrence probabilities for each failure occurrence location are displayed in descending order of probability.
[Form 3]
As shown in FIG. 7, first, the monitoring parameter data necessary for calculating the feature amount is obtained by referring to the latest machine information stored in the machine information storage unit 22 for the image forming apparatus 10 to be predicted for failure. (Step 110), and the feature amounts are calculated by the same method as the failure sign determination model generation unit 28 (step 112).

  Next, a failure type determination probability distribution is acquired from the failure sign determination model generation unit 28 (step 114), and for each failure type, the probability that that type of failure will occur in the near future of the failure prediction target image forming apparatus 10 is obtained. (Predictive probability of occurrence) is calculated (step 116).

  Also, a failure location determination probability distribution is acquired from the failure sign determination model generation unit (step 120), and for each failure occurrence location, the probability that the failure at that location will occur in the near future of the failure prediction target image forming apparatus 10 (Predictive probability of occurrence) is calculated (step 122).

  Thereafter, a preparatory process is performed in which the occurrence predictor probability calculated for each failure type and the occurrence predictor probability calculated for each failure occurrence location are classified and notified for each failure type (step 126).

  “Preparation for notification” in step 126 indicates that notification may not be made depending on notification and maintenance determination described later, but a maintenance request is not required, but only notification may be executed.

  In order to classify the occurrence probabilities for each failure occurrence location by failure type, for example, a correspondence table in which a failure type and a failure occurrence location are associated with each other may be prepared in advance and performed according to the correspondence table.

  Note that all the calculated occurrence probabilities may be notified, or only those exceeding a predetermined threshold value may be notified. Further, when notifying a plurality of occurrence predictor probabilities, it is preferable to notify them in descending order of probability. The notification of the occurrence predictor probability is performed by, for example, displaying and outputting with a display device provided in the maintenance information input terminal 12.

FIG. 9C illustrates an example in which the occurrence predictor probability for each failure type and the occurrence predictor probability for each failure location are classified by failure type and notified in a list format. The occurrence predictor probability for each failure type Are displayed in descending order of probability, and the occurrence predictor probabilities of the fault occurrence locations corresponding to the types of failures are displayed in descending order of probability.
[Form 4]
As shown in FIG. 8, first, the monitoring parameter data necessary for calculating the feature amount with reference to the latest machine information stored in the machine information storage unit 22 for the image forming apparatus 10 to be predicted for failure. (Step 130), and the feature amounts are calculated by the same method as the failure sign determination model generation unit 28 (step 132).

  Next, a failure type determination probability distribution is acquired from the failure sign determination model generation unit 28 (step 134), and for each failure type, the probability that that type of failure will occur in the near future of the failure prediction target image forming apparatus 10. (Predictive probability of occurrence) is calculated (step 136).

  Also, a failure location determination probability distribution is acquired from the failure sign determination model generation unit 28 (step 138), and for each failure occurrence location, a failure at that location will occur in the near future to the failure prediction target image forming apparatus 10. A probability (probability of occurrence) is calculated (step 140).

  Next, each occurrence predictor probability calculated for each failure occurrence location is compared with a predetermined threshold (step 142), and for the failure occurrence location where the probability is equal to or greater than the threshold, the failure type having the location as the main cause is identified. Then, correction is performed to increase the probability of occurrence of the failure type by a predetermined rate (step 144). The failure type may be specified in accordance with, for example, a correspondence table between the failure type and the location where the failure has occurred.

  After the processing (step 142, step 144) for correcting the occurrence predictor probability for each failure type is performed for all the failure occurrence locations (affirmative determination in step 146), the occurrence predictor probability (before correction and After the correction), a preparatory process is performed in which the occurrence predictor probability calculated for each failure occurrence location is classified and notified for each failure type (step 148).

  “Preparation for notification” in step 148 indicates that notification may not be performed depending on notification and determination of whether maintenance is possible, which will be described later. However, even if a maintenance request is not required, only notification may be executed.

  In order to classify the occurrence probabilities for each failure location by failure type, for example, it may be performed according to a correspondence table between failure types and failure occurrence locations.

  Note that all the calculated occurrence probabilities may be notified, or only those exceeding a predetermined threshold value may be notified. Further, when notifying a plurality of occurrence predictor probabilities, it is preferable to notify them in descending order of probability.

  FIG. 9D shows an example in which the occurrence predictor probability (before and after correction) for each failure type and the occurrence predictor probability for each failure location are classified by failure type and notified in a list format. The occurrence predictor probabilities for each type of failure are displayed in descending order of the probability after correction, and the occurrence predictor probabilities for the failure occurrence locations corresponding to each type of failure are displayed in descending order of probability.

  In the failure prediction system according to the present embodiment, the failure predictor determination model generation unit 28 of the management device 14 has a failure in the image forming device 10 for each of a plurality of types of feature quantities indicating the characteristics of the internal state of the image forming device 10. A failure determination probability distribution indicating the distribution of the occurrence frequency of the feature amount when it occurs and a failure determination probability distribution indicating the distribution of the occurrence frequency of the feature amount when no failure has occurred in the image forming apparatus 10 are stored. The failure occurrence prediction unit 30 of the management device 14 calculates the value of each feature amount from the latest machine information stored in the machine information storage unit 22 for the image forming apparatus 10 that is the target of failure prediction, and the calculation Based on the value of each feature amount and the failure probability distribution for each feature amount accumulated in the failure sign determination model generation unit 28, the failure predictor probability in the image forming apparatus 10 that is the target of the failure calculation is calculated. Was to so that.

  That is, the most recent feature value in the failure prediction target image forming apparatus 10 is calculated, and the probability that a failure has occurred in the past case where a combination of values similar to the calculated feature value is obtained, and Considering the probability that no failure has occurred, the probability that a failure will occur in the near future of the failure prediction target image forming apparatus 10 is calculated, so the failure occurrence probability can be accurately calculated.

(Maintenance judgment index generation processing)
FIG. 10 is a control flowchart showing a maintenance determination index generation process in the maintenance determination index generation unit 34 of the notification unit 32 (see FIG. 1).

  In step 150, a model generation target failure occurrence example is extracted from the maintenance information storage unit 24, and then the process proceeds to step 152 to monitor parameter data for the period ΔT3 shown in FIG. Is acquired from the machine information storage unit 22, and the process proceeds to step 154.

  In step 154, various feature amounts are calculated based on each monitoring parameter data, and then the process proceeds to step 156 to calculate an occurrence predictor probability of a failure type using a failure predictor determination model.

  In the next step 158, a maintenance determination parameter (toner consumption per unit number in this embodiment) is extracted, and then the process proceeds to step 160 where the predictive probability is equal to or higher than a predetermined criterion that a failure has occurred. Toner consumption data per unit number is extracted for the period ΔT4 (see FIG. 2) including the date (determination date).

  In the next step 162, the maintenance determination parameters (extracted toner consumption data group per unit number) included in the period ΔT2 starting from the failure occurrence date and before the starting point and other than the period ΔT2 Categorize into things.

  In the next step 164, a threshold value is obtained that is obtained from the calculation result of the expression (2) and has a maximum degree of separation between the maintenance determination parameters included in the period ΔT2 and the maintenance determination parameters other than the period ΔT2. And the routine ends.

(Maintenance target judgment processing)
FIG. 11 is a control flowchart showing a maintenance target determination process in the maintenance target determination unit 36 of the notification unit 32 (see FIG. 1). In this maintenance target determination process, the maximum value (threshold value) of the degree of separation stored by the maintenance determination index generation process is used.

  In step 170, the latest monitoring parameter data is acquired from the image forming apparatus 10 to be monitored, and then the process proceeds to step 172 to calculate various feature amounts from each monitoring parameter data, and the process proceeds to step 174.

  In step 174, the failure predictor probability is calculated using the failure predictor determination model, and in the next step 176, it is determined whether or not the calculated predictor probability is equal to or greater than a predetermined criterion for failure occurrence. .

  If a negative determination is made in step 176, it is determined that no failure has occurred, and this routine ends.

  If an affirmative determination is made in step 176, it is determined that a failure has occurred, and the process proceeds to step 178.

  In step 178, the data of the maintenance determination index is extracted for the latest period ΔT4 including the determination date, the average value is calculated, and the process proceeds to step 180.

  In step 180, it is determined whether or not the average value calculated in step 178 is greater than or equal to the threshold value stored in step 164 of FIG.

  If the determination in step 180 is negative, it is determined that the image forming apparatus 10 to be monitored is outputting mainly the text image, and the user who manages the image forming apparatus 10 to be monitored is notified of the occurrence of the failure. Neither does it request maintenance by a service engineer. It should be noted that the notification of the occurrence of the failure may be made and the maintenance by the service engineer may not be requested.

  Whether or not to notify may depend on the user's selection, for example.

  If the determination in step 180 is affirmative, it is determined that the image forming apparatus 10 to be monitored is outputting mainly the photographic image, and the process proceeds to step 182 where the failure prediction notification unit 40 (see FIG. 1). The user who manages the image forming apparatus 10 to be monitored is notified of the occurrence of a failure and requested to be maintained by a service engineer.

  The management device 14 in the failure prediction system of the present embodiment includes a CPU (Central Processing Unit) that performs various arithmetic processes, a RAM (Random Access Memory) that is a work area of the CPU, and a ROM (Read) that records a basic control program. Main memory such as Only Memory), auxiliary storage such as HDD (Hard Disk Drive) that stores various programs and data, display device for displaying and outputting various information, and operations used for input operations by the operator Consists of a computer having hardware resources such as an input / output I / F that is an interface with an input device such as a button or a touch panel, and a communication I / F that is an interface for performing wired or wireless communication with other devices. Has been.

  In addition, the program according to the present embodiment may be a computer according to the present embodiment, for example, in a format read from an external storage medium such as a CD-ROM storing the program or received via a communication network. Set to

  Furthermore, the present invention is not limited to a mode in which each functional unit is realized by a software configuration as in the present embodiment, and each functional unit may be realized by a dedicated hardware module.

  In this embodiment, the period ΔT shown in FIG. 2 is about 1 year, the period ΔT1 is about the first half of the period ΔT, ΔT2 is about 1 to 2 weeks, ΔT3 is about the last half of the period ΔT, and ΔT4 is Although about 1 to 2 weeks, these periods are not limited. Each period is a guideline and is described as “...”, But in other words, it does not affect the data collection effect even if the period is not clearly set.

  Further, for example, since data collection is so-called past data, there may be no history when the apparatus is introduced. Therefore, when the apparatus is introduced, data obtained at the start stage may be used for an apparatus that has been subjected to an endurance test or the like before the apparatus is introduced.

  Furthermore, in the present embodiment, the threshold value for determining whether maintenance is necessary is fixed, but depending on the user, the text image is the mainstream, but the photographic image increases, or the photographic image is the mainstream. It is expected that there will be an increase in the number of text images. Therefore, the threshold value for determining the necessity of maintenance may be updated regularly or irregularly.

P Recording sheet 10 Image forming apparatus 12 Maintenance information input terminal 14 Management device 16 Data acquisition unit 18 Machine information collection unit 20 Maintenance information collection unit 22 Machine information storage unit 24 Maintenance information storage unit 26 Model generation unit 28 Fault sign determination model generation unit 30 Failure occurrence prediction unit 32 Notification unit 34 Maintenance determination index generation unit 36 Maintenance target determination unit 40 Failure predictor notification unit

Claims (5)

Prediction means for predicting the occurrence of a failure based on the probability of occurrence of a failure calculated by classifying information related to the maintenance history and processing history of the monitored device into normal time and abnormal time,
A determination unit that determines a determination index serving as an index for determining whether or not the maintenance corresponding to the occurrence of the failure is necessary based on output information regarding the image output by the monitoring target device;
Based on the determination index determined by the determination unit, a determination unit that determines whether or not maintenance is necessary for a device in which the occurrence of a failure is predicted by the prediction unit;
A failure prediction management apparatus. When the occurrence of a failure is predicted by the prediction unit, the management unit that manages the corresponding monitoring target device further includes a notification unit that notifies the occurrence of the failure,
The failure prediction management apparatus according to claim 1, wherein it is possible to select to always notify when a failure occurrence is predicted or to depend on a determination result of the determination unit. The prediction means is
The first distribution indicating the distribution of the occurrence frequency of the feature amount classified at the time of the abnormality and the second distribution indicating the distribution of the occurrence frequency of the feature amount classified at the time of the normal are stored in the monitoring target device. The failure prediction management apparatus according to claim 1, wherein a failure occurrence probability is calculated by acquiring a current value of each feature amount and collating with the first distribution and the second distribution. The output information is at least one piece of numerical information of toner consumption per unit number of sheets, image forming area, or image density information;
The threshold value which is the tolerance | permissibility of the image quality requested | required of the apparatus of monitoring object with respect to the said numerical information is defined, The said threshold value is used as said determination parameter | index. Failure prediction management device. On the computer,
Classify the information related to the maintenance history and processing history of monitored devices into normal and abnormal conditions,
Storing a first distribution indicating a distribution of occurrence frequencies of feature quantities classified at the time of abnormality and a second distribution indicating a distribution of occurrence frequencies of feature quantities classified at the time of normality;
Obtaining the current value of each feature quantity in the device to be monitored, calculating the occurrence probability of the failure by collating with the first distribution and the second distribution,
Based on the output information related to the image output by the monitoring target device, determine a determination index that is an index for determining whether maintenance is required in response to the occurrence of a failure,
Based on the determined determination index, determine whether maintenance is necessary for a device that is predicted to have a failure,
A failure prediction management program that makes things happen.