JP2012037991A - Prediction device, prediction system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable an accurate prediction of an occurrence of an abnormality in a monitor object device.SOLUTION: In a prediction device S of this embodiment, a device information collection part 11 acquires status information from an image formation device P which is an object for an abnormality prediction; a distance calculation part 13 calculates the distances to a normal space and an abnormal space for each abnormality of the prediction object, based on the status information acquired by the device information collection part 11; a distance variation calculation part 14 calculates the variation of the distance to the normal space and the variation of the distance to the abnormal space for each abnormality of the prediction object, base on the calculation results by the distance calculation 13; and an abnormal occurrence prediction part 15, based on the calculation results by the distance calculation part 13 and the distance variation calculation part 14, specifies the abnormality which can be assumed to highly occur in the near future in the image formation device P being an object for the abnormality prediction on the basis of predetermined conditions, and predicts the occurrence time of the abnormality.

Description

  The present invention relates to a prediction device, a prediction system, and a program.

Various inventions have been proposed regarding prediction of occurrence of an abnormality in a monitoring target device (for example, an image forming device).
For example, an invention is proposed in which multiple types of information are periodically acquired from an image forming apparatus, the Mahalanobis distance with respect to a normal state is calculated as an index value, and if the index value exceeds a threshold, it is determined that the possibility of occurrence of an abnormality is high (See Patent Document 1).
For example, a plurality of normal index information having different values are prepared as indicators of the normal state of the subject to be examined, and the abnormality value is determined by calculating the index value by switching the normal index information according to specific information such as environmental information. An invention to be performed and an invention to change the determination frequency of the presence / absence of abnormality according to the rate of increase of the index value have been proposed (see Patent Document 2).

JP 2005-017874 A JP 2005-258384 A

  An object of this invention is to propose the technique which can perform the prediction about generation | occurrence | production of abnormality in the monitoring object apparatus with sufficient accuracy.

  According to the first aspect of the present invention, the abnormality occurs in the acquisition unit that acquires the state detected by the monitoring target device for the monitoring item related to the abnormality to be predicted, and the state acquired by the acquisition unit. The occurrence of the abnormality in the monitoring target device based on the calculation means for calculating the distance to the abnormal space that is predetermined as the space that represents the state of the situation, and the distance to the abnormal space calculated by the calculation means And a prediction device for prediction.

  According to a second aspect of the present invention, in the prediction apparatus according to the first aspect, the prediction unit is configured to change a distance with respect to the abnormal space respectively calculated for states detected at different times in the monitoring target apparatus. Based on this, the time when the abnormality occurs in the monitoring target device is predicted.

  According to a third aspect of the present invention, in the prediction apparatus according to the second aspect, the calculation unit preliminarily represents a space representing the state when the abnormality has not occurred in the state acquired by the acquisition unit. Further calculating a distance to the determined normal space, and the predicting means, when the distance to the abnormal space calculated by the calculating means is shorter than the distance to the normal space calculated by the calculating means The time when the abnormality occurs in the target device is predicted.

  According to a fourth aspect of the present invention, in the prediction device according to any one of the first to third aspects, the monitoring item is a monitoring item other than the monitoring item related to the abnormality to be predicted, and the abnormality occurs. A monitoring item having a commonality that is greater than or equal to a reference in a state of change detected in some of the monitoring target devices among a plurality of monitoring target devices having a device attribute that identifies the part of the monitoring target devices An additional means for adding to a monitoring item related to the abnormality of the monitoring target device is provided.

  The present invention according to claim 5 includes a monitoring target device and a prediction device, and the monitoring target device includes a detection unit that detects a state of a monitoring item related to an abnormality of the prediction target, and the prediction When the abnormality has occurred in the acquisition unit that acquires the state of the monitoring item related to the abnormality detected in the monitoring target device from the monitoring target device, and the state acquired by the acquisition unit A calculation unit that calculates a distance to an abnormal space that is predetermined as a space that represents a state of the condition, and a prediction that predicts the occurrence of the abnormality in the monitoring target device based on the distance to the abnormal space calculated by the calculation unit A prediction system comprising: means.

  The present invention described in claim 6 provides the computer with an acquisition function for acquiring a state detected by the monitoring target device for a monitoring item related to the abnormality to be predicted, and for the state acquired by the acquisition function. Based on a calculation function for calculating a distance to an abnormal space that is predetermined as a space that represents a state in the case of occurrence of the error, and the distance to the abnormal space calculated by the calculation function, the abnormality in the monitoring target device This is a program for realizing a prediction function for predicting the occurrence of the occurrence of the problem.

  According to the present invention described in claims 1, 5, and 6, regarding the prediction of the occurrence of abnormality in the monitoring target device, the occurrence of the abnormality of the prediction target can be predicted with higher accuracy than when the present invention is not applied. .

  According to the second aspect of the present invention, it is possible to predict the time when the abnormality to be predicted occurs more accurately than when the present invention is not applied.

  According to the third aspect of the present invention, it is possible to predict when the abnormality to be predicted occurs when the possibility of the occurrence of the abnormality has increased to some extent.

  According to the fourth aspect of the present invention, the monitoring items related to the abnormality of the prediction target can be made different according to the device attributes such as the installation environment (for example, the region) and the model of the monitoring target device.

It is a figure which illustrates the functional block of the prediction system which concerns on one Embodiment of this invention. It is a figure which illustrates the flow of the prediction process by the prediction system which concerns on one Embodiment of this invention. It is a figure which illustrates transition of the distance with respect to the normal space calculated in the prediction system concerning one embodiment of the present invention, and the distance to abnormal space. It is a figure which illustrates the main hardware constitutions of the computer which operate | moves as a prediction apparatus in the prediction system which concerns on one Embodiment of this invention.

An embodiment of the present invention will be described with reference to the drawings.
FIG. 1 illustrates functional blocks of a prediction system according to an embodiment of the present invention. The prediction system of the present example includes an image forming apparatus P that forms an image on a medium such as paper, a maintenance information input terminal T that is used by an administrator or maintenance person of the image forming apparatus P, and an abnormality in the image forming apparatus P. A prediction device S for predicting the occurrence. In the example of FIG. 1, two image forming apparatuses P and one maintenance information input terminal T are shown, but the number of these is arbitrary.

  The image forming apparatus P is an example of a monitoring target apparatus that is a target of abnormality prediction by remote monitoring, and has functions such as a printing apparatus (printer), a copying apparatus (copying machine), a facsimile apparatus, and a printing / copying / facsimile function. This includes multifunction devices in preparation for. The monitoring target apparatus is not limited to the image forming apparatus P, and can be various apparatuses that can communicate with the prediction apparatus S wirelessly or by wire. The main target is a device that is required to respond quickly in the event of a failure, such as a device that does not have two standby systems) or a device that is installed assuming normal operation.

  The image forming apparatus P of this example includes a plurality of detection units (sensors) for detecting states of various monitoring items such as temperature and humidity inside the apparatus, and is intermittent (for example, predetermined). At every time interval, every time a predetermined number of image forming processes are executed, when the apparatus switches to a predetermined operation mode, etc., the status of each monitoring item is detected, and each monitoring item is detected. The status information indicating the status (detected value) is stored in the storage unit inside the apparatus. The state information accumulated in the storage unit inside the apparatus is transmitted to the prediction apparatus S by wireless or wired communication at predetermined time intervals or in response to an instruction from the outside. Examples of the instruction from the outside include an instruction received by the image forming apparatus P itself and an instruction transmitted to the image forming apparatus P from another apparatus such as the prediction apparatus S or the maintenance information input terminal T.

  The maintenance information input terminal T transmits information to the prediction device S by wireless or wired communication according to an operation received from the operator, or displays the display based on the information received from the prediction device S. For example, output is performed by a display device provided in the terminal T. The maintenance information input terminal T of the present example was performed on the image forming apparatus P from a maintenance person who actually visited the place where the image forming apparatus P was installed and performed maintenance work, or a person who received the report. The input of maintenance information regarding the maintenance work is received and transmitted to the prediction device S. In addition, information related to a prediction result related to the occurrence of a failure in the image forming apparatus P is received from the prediction apparatus S and displayed on the display apparatus.

  The prediction device S includes a device information collection unit 11, a normal / abnormal space generation unit 12, a distance calculation unit 13, a distance change calculation unit 14, an abnormality occurrence prediction unit 15, a state information storage unit 16, a maintenance information storage unit 17, a space It has functional units such as the information storage unit 18 and the distance information storage unit 19, and it is possible to identify an abnormality that is highly likely to occur in the image forming apparatus P from among a plurality of abnormalities to be predicted and the occurrence thereof. Predict the time.

  The apparatus information collection unit 11 acquires the state information transmitted from the image forming apparatus P and stores it in the state information storage unit 16. In the status information of this example, an item ID for identifying a monitoring item, a status (detected value) detected for the monitoring item, a date and time when the status of the monitoring item is detected, and the like are recorded. A device ID that uniquely identifies the image forming apparatus P that has become, a model ID that identifies the model, and the like are associated with each other.

In addition, the device information collection unit 11 acquires maintenance information transmitted from the maintenance information input terminal T and stores it in the maintenance information storage unit 17. The maintenance information in this example records the maintenance contents indicating the contents of the maintenance work, the date and time when the maintenance work was performed, the abnormality to be removed by the maintenance work, and the image formation that is the target. A device ID that uniquely identifies the device P, a model ID that identifies a model, and the like are associated with each other.
In this example, maintenance information received from a maintenance person or the like is obtained by communication from the maintenance information input terminal T, but maintenance information may be obtained by other configurations. It is possible to use a configuration in which the input of maintenance information is received by the image forming apparatus P that has been acquired and acquired through communication, or a configuration in which the input of maintenance information is directly received by the prediction apparatus S.

  Based on the state information stored in the state information storage unit 16 and the maintenance information stored in the maintenance information storage unit 17, the normal / abnormal space generation unit 12 generates an abnormality for each of a plurality of abnormalities to be predicted. A normal space representing a state when no abnormality has occurred and an abnormal space representing a state when an abnormality has occurred are generated, and spatial information representing each space is stored in the spatial information storage unit 18. The normal space and the spatial information of the abnormal space generated for each abnormality to be predicted are used in the subsequent processing to predict the occurrence of the corresponding abnormality.

Spatial information generation processing executed in the normal / abnormal space generation unit 12 will be described.
In the normal / abnormal space generation unit 12 of the present example, the following processing is performed with each of a plurality of abnormalities to be predicted being sequentially processed.
First, before and after the maintenance work in each of the plurality of image forming apparatuses P, among the various monitoring items whose states are detected in the plurality of image forming apparatuses P on which the maintenance work for eliminating the abnormality to be processed is performed. In this case, one or more monitoring items in which changes characteristic to the state (detection value) (for example, the difference between the states before and after the maintenance work is greater than or equal to the threshold value) are identified as monitoring items related to the abnormality of the processing target. .

Then, for each identified monitoring item, a statistical value (for example, an average) relating to the state before the maintenance work is performed is calculated, and a vector space having the calculated statistical value of each monitoring item as an element Is generated as an abnormal space for the abnormality to be processed. Further, a statistical value (for example, an average) related to the state after the maintenance work is performed for each specified monitoring item is calculated, and a vector space having the statistical value of the state of each monitoring item as an element is calculated. Spatial information is generated as a normal space for the abnormality to be processed.
After that, each spatial information is associated with an abnormality ID for identifying abnormality and stored in the spatial information storage unit 18.

  As a result, in the normal space and the abnormal space in this example, the monitoring item that is estimated to have changed in relation to the target abnormality is used as a component of the space, while it is estimated that it is unrelated to the abnormality. The monitored items are excluded from the space components. For this reason, the monitoring items relating to the components of the normal space and the abnormal space in this example correspond to the target abnormality.

  In this example, the image forming apparatus P prepared for the test is subjected to a stress test or the like in advance so as to intentionally generate a failure, and spatial information generation processing is performed based on the information collected at that time. The spatial information generated as a result is stored in the spatial information storage unit 18 as an initial value, and thereafter, based on information collected for a plurality of image forming apparatuses P actually operated in the market in advance. Spatial information generation processing is re-executed every predetermined period (for example, January) to update the spatial information in the spatial information storage unit 18. In addition, triggered by other conditions, such as a configuration that updates spatial information in response to the arrival of a predetermined date and time, a configuration that updates spatial information in response to an instruction input by a system administrator, etc. The spatial information may be updated.

  Based on the state information stored in the state information storage unit 16 and the spatial information stored in the spatial information storage unit 18, the distance calculation unit 13 relates to each abnormality for each of a plurality of abnormalities to be predicted. The distance of the normal space and the distance of the abnormal space with respect to the state detected by the image forming apparatus P that is the target of abnormality prediction for the monitoring item to be calculated are calculated, and the distance information representing each distance is stored in the distance information storage unit 19.

The distance calculation process executed in the distance calculation unit 13 will be described.
In the distance calculation unit 13 of this example, the following processing is performed with each of a plurality of abnormalities to be predicted being sequentially processed.
First, for each monitoring item related to a normal space and an abnormal space component regarding the abnormality to be processed, a corresponding state (detected value) included in the state information acquired from the image forming apparatus P that is the target of abnormality prediction is displayed. Each is extracted and vectorized.
Next, the Mahalanobis distance between the extracted vector and the normal space for the abnormality to be processed is calculated as the distance to the normal space. Further, the Mahalanobis distance between the extracted vector and the abnormal space for the abnormality to be processed is calculated as a distance to the abnormal space.
Thereafter, the distance information representing each distance is stored in the distance storage unit 19 in association with an apparatus ID for identifying the image forming apparatus P, an abnormality ID for identifying an abnormality, a date and time when the distance is calculated, and the like.

As a result, the distance storage unit 19 of the present example stores the distance to the normal space and the distance to the abnormal space calculated for the state detected at different time points for each image forming apparatus P and each abnormality to be predicted. Information is stored.
In this example, every time the status information is acquired from the image forming apparatus P and stored in the status information storage unit 16, the distance is calculated for the status information (that is, the latest status information). The distance may be calculated collectively for a plurality of different state information.
In this example, the Mahalanobis distance is used as the distance to the normal space and the abnormal space, but a distance calculated by another method may be used.

  Based on the distance information stored in the distance information storage unit 19, the distance change amount calculation unit 14 is configured for each of a plurality of abnormalities to be predicted with respect to the normal space related to the image forming apparatus P that is the target of abnormality prediction. A change between the latest distance and the previous distance, and a change between the latest distance and the previous distance with respect to the abnormal space are calculated.

  The abnormality occurrence prediction unit 15 is close to the image forming apparatus P that is the target of abnormality prediction based on the calculation result by the distance calculation unit 13 (the contents stored in the distance information storage unit 19) and the calculation result by the distance change calculation unit 14. An abnormality that is estimated to have a high possibility of occurring in the future is specified based on a predetermined condition, and the time when the abnormality occurs is predicted.

The abnormality occurrence prediction process executed in the abnormality occurrence prediction unit 15 will be described.
The abnormality occurrence prediction unit 15 of the present example is based on the following (Condition 1) to (Condition 3), and in the near future in the image forming apparatus P that is the target of abnormality prediction from among a plurality of abnormalities that are prediction targets. Identify anomalies that are likely to occur.

(Condition 1) If there is one abnormality that satisfies “distance to normal space> distance to abnormal space”, the abnormality is selected.
(Condition 2) When there are a plurality of abnormalities satisfying “distance to normal space> distance to abnormal space”, the change between the latest distance to the abnormal space and the previous distance (however, the change in the direction approaching the abnormal space) is Pick the biggest anomaly.
(Condition 3) If there is no abnormality such as “distance to normal space> distance to abnormal space”, the difference between the latest distance to normal space and the previous distance, and the latest distance to abnormal space and the previous distance Based on the change, it is determined whether there is an abnormality that satisfies “distance to normal space> distance to abnormal space” within a predetermined period, and the corresponding abnormality is selected. When there are a plurality of corresponding abnormalities, the abnormality having the largest change between the latest distance to the abnormal space and the previous distance (however, the change in the direction approaching the abnormal space) is selected.

  Thereafter, the anomaly selected based on (Condition 1) to (Condition 3) is identified as an anomaly that is estimated to be highly likely to occur in the near future. Predict when it will occur. In this example, the time when the distance to the abnormal space is less than a predetermined threshold (or less than the threshold) is predicted as the time when the abnormality occurs.

  In addition to predicting the timing of occurrence of an abnormality as described above, an abnormality that is highly likely to occur in the near future (or has already occurred) may be predicted. Specifically, for example, for each of a plurality of anomalies to be predicted, it is determined whether the latest distance to the anomaly space is less than a predetermined threshold (or less than the threshold), and less than the threshold (or less than the threshold). It is predicted that an abnormality determined to be an abnormality that is likely to occur (or has already occurred) in the near future.

FIG. 2 illustrates a flow of prediction processing by the prediction device S of this example.
In the prediction apparatus S of this example, the apparatus information collection unit 11 acquires state information from the image forming apparatus P that is a target of abnormality prediction (step S11), and the distance calculation unit 13 is acquired by the apparatus information collection unit 11. Based on the state information, the normal space and the distance to the abnormal space (Mahalanobis distance in this example) are calculated for each abnormality to be predicted (step S12), and the distance change calculation unit 14 calculates the calculation result by the distance calculation unit 13 Based on the above, a change in distance to the normal space and a change in distance to the abnormal space are calculated for each abnormality to be predicted (step S13).

  After that, the abnormality occurrence prediction unit 15 compares the distance to the normal space calculated for each abnormality of the prediction target by the distance calculation unit 13 and the distance to the abnormal space (step S14), and for all the abnormalities to be predicted, It is determined whether or not the condition “distance to normal space <distance to abnormal space” is satisfied (step S15).

  If a positive determination result (YES) is obtained in step S15, it is further determined whether or not there is an abnormality that satisfies the condition “distance to normal space> distance to abnormal space” within a predetermined period. Determination is made (step S16). If a positive determination result (YES) is obtained in this determination, the time when the abnormality occurs is predicted based on the transition of the distance to the abnormal space (step S19), and the negative determination result (NO) If obtained, the process is terminated.

  On the other hand, when a negative determination result (NO) is obtained in step S15, it is determined whether or not there is one abnormality that satisfies “distance to normal space> distance to abnormal space” (step S17). . If a positive determination result (YES) is obtained in this determination, the time when the abnormality occurs is predicted based on the transition of the distance to the abnormal space (step S19), and the negative determination result (NO) Is obtained, an anomaly having the largest change in the distance to the anomalous space is selected (step S18), and the time when the anomaly occurs is predicted based on the transition of the distance to the anomaly space (step S19). .

  The result of performing the prediction process is transmitted to the maintenance information input terminal T, and is displayed and output by the display device of the maintenance information input terminal T. Note that the output of the prediction result may be performed using another output method such as, for example, sending an e-mail to a predetermined transmission destination (such as an administrator or maintenance staff of the image forming apparatus P). .

In addition, said prediction process is only an example and the some modification is shown below.
For example, the occurrence time may be predicted for all the abnormalities where “distance to normal space> distance to abnormal space”. Note that the occurrence timing may be predicted only once when “distance to normal space> distance to abnormal space” or “distance to normal space> distance to abnormal space” occurs. Time prediction may be repeated multiple times.
For example, the occurrence time may be predicted by selecting a predetermined number of abnormalities in the order of high probability of occurrence of abnormalities (in order of the distance from the abnormal space).
When the occurrence times are predicted for a plurality of abnormalities, prediction results may be output for a predetermined number of abnormalities in the order of occurrence time.

  FIG. 3 illustrates the transition of the distance to the normal space and the distance to the abnormal space calculated in the prediction system of this example. The vertical axis represents the distance calculated for each abnormality, and the horizontal axis represents the passage of time. Further, α is indicated by a threshold regarding the distance to the abnormal space, and it is predicted that an abnormality will occur when the distance to the abnormal space becomes less than α (or less than α).

  3A illustrates the distance LA1 to the normal space and the distance LA2 to the abnormal space calculated for the abnormality A, and FIG. 3B illustrates the distance LB1 to the normal space calculated for the abnormality B. 3B illustrates the distance LB2 to the abnormal space, and FIG. 3C illustrates the distance LC1 to the normal space calculated for the abnormality C and the distance LC2 to the abnormal space.

According to FIG. 3, for abnormality A, the distance LA1 to the normal space gradually increases (shifts away from the normal space), but the distance LA2 to the abnormal space also gradually increases (shifts away from the abnormal space). It can be seen that abnormality A does not occur for the time being.
On the other hand, for the anomaly B, the distance LB1 with respect to the normal space gradually increases (shifts away from the normal space), while the distance LB2 with respect to the anomalous space gradually decreases (shifts toward the anomaly space). It can be seen that the abnormality B is approaching. As for the abnormality C, as with the abnormality B, the distance LC1 to the normal space gradually increases (shifts away from the normal space), while the distance LC2 to the abnormal space gradually decreases (changes closer to the abnormal space as well). It can be seen that the occurrence of abnormality C is approaching. And according to the decreasing degree of the distance with respect to abnormal space, it turns out that the number of days until abnormality B becomes less than (or less than or equal to) α is short and the degree of urgency is high.

  Here, in the prediction system described so far, the monitoring items common to all the image forming apparatuses P that are targets of abnormality prediction are used as the monitoring items related to the normal space for each abnormality to be predicted and the components of the abnormal space. However, it is also possible to additionally use specific monitoring items according to the device attributes such as the installation environment (for example, the region) and model of the image forming apparatus P.

An example of a configuration that additionally uses unique monitoring items according to the device attributes of the image forming apparatus P will be described.
For example, the normal / abnormal space generation unit 12 performs the following processing with each of a plurality of abnormalities to be predicted in turn as processing targets.
First, some image formation is performed from among various monitoring items (excluding common monitoring items) whose states are detected in a plurality of image forming apparatuses P on which maintenance work for eliminating the abnormality to be processed is performed. In the device P, one or more monitoring items in which a characteristic change in the state (detected value) before and after the maintenance work (for example, the difference in the state before and after the maintenance work is greater than or equal to the threshold value) are identified. The identified monitoring item is a unique monitoring item that is additionally applied to the group of image forming apparatuses P having the same apparatus attribute as the apparatus attribute shared by some of the image forming apparatuses P.

Next, for each of the common monitoring items applied to all the image forming apparatuses P and the specific monitoring items specified for some of the image forming apparatuses P, the statistical values relating to the state before the maintenance work is performed ( For example, an average) is calculated, and space information is generated using a vector space having the calculated statistical value of each monitoring item as an element as an abnormal space for the abnormality to be processed. Further, a statistical value (for example, an average) related to the state after the maintenance work is performed for each specified monitoring item is calculated, and a vector space having the statistical value of the state of each monitoring item as an element is calculated. Spatial information is generated as a normal space for the abnormality to be processed.
Thereafter, an abnormality ID for identifying an abnormality, an attribute ID for identifying a device attribute, and the like are associated with each spatial information and stored in the spatial information storage unit 18.

  In the prediction of the occurrence of abnormality, the device attribute of the image forming apparatus P that is the target of abnormality prediction is referred to, and the normal space associated with the corresponding device ID and the space information of the abnormal space are normal / abnormal space. If it exists in the generation unit 12, the distance to these spaces is calculated and predicted.

  As a method for specifying a specific monitoring item, for example, a method using cluster analysis can be cited. That is, a data group having the state of each monitoring item (excluding a common monitoring item) as an element is formed for each image forming apparatus P, and the data group for each image forming apparatus P is set to a predetermined classification level (classification granularity). ) To perform clustering processing to classify into multiple sets, and select the set of monitoring items that contributed to the classification that has a large change in state before and after maintenance work from among the multiple classified sets. This is a monitoring item unique to the image forming apparatus P group. When a plurality of sets are selected, they are added as unique monitoring items for each image forming apparatus P group corresponding to each set.

FIG. 4 illustrates a main hardware configuration of a computer that operates as the prediction device S in the prediction system of this example.
In this example, a main memory such as a CPU (Central Processing Unit) 31 that performs various arithmetic processes, a RAM (Random Access Memory) 32 that is a work area of the CPU 31, and a ROM (Read Only Memory) 33 that records basic control programs. Device, auxiliary storage device (for example, magnetic disk such as HDD (Hard Disk Drive), rewritable nonvolatile memory such as flash memory, etc.) 34 for storing programs and various data according to an embodiment of the present invention, various types Communication between the display device for displaying information and the input / output I / F 35 that is an interface with an input device such as an operation button or a touch panel used for an input operation by an operator, or other devices by wire or wireless Interface to do It is constituted by the communication I / F 36, a computer having the hardware resources and the like.
Then, the program according to the embodiment of the present invention is read from the auxiliary storage device 34 and the like, loaded into the RAM 32, and executed by the CPU 31, thereby causing each function of the prediction device S according to the embodiment of the present invention to be executed by a computer. Realized above.

In this example, each function unit of the prediction device S is provided in one computer, but each function unit may be provided in a distributed manner in a plurality of computers.
In addition, the program according to the embodiment of the present invention may be a computer according to the present example, for example, in a format read from an external storage medium such as a CD-ROM storing the program or a format received via a communication line. Set to
Moreover, it is not restricted to the aspect which implement | achieves each function part by software configuration like this example, You may make it implement | achieve each function part with a dedicated hardware module.

P: Image forming apparatus, T: Maintenance information input terminal T, S: Prediction apparatus,
11: Device information collection unit, 12: Normal / abnormal space generation unit, 13: Distance calculation unit, 14: Distance change calculation unit, 15: Abnormality occurrence prediction unit, 16: Status information storage unit, 17: Maintenance information storage unit 18: Spatial information storage unit 19: Distance information storage unit

Claims (6)

An acquisition means for acquiring a state detected by the monitoring target device with respect to a monitoring item related to the abnormality to be predicted;
For the state acquired by the acquisition unit, a calculation unit that calculates a distance to an abnormal space that is predetermined as a space that represents the state when the abnormality occurs;
Prediction means for predicting the occurrence of the abnormality in the monitoring target device based on the distance to the abnormal space calculated by the calculation means;
A prediction apparatus comprising: The predicting means predicts a time when the abnormality occurs in the monitoring target device based on a transition of a distance with respect to the abnormal space respectively calculated for states detected at different times in the monitoring target device;
The prediction apparatus according to claim 1. The calculation means further calculates a distance with respect to a normal space that is predetermined as a space representing a state when the abnormality has not occurred with respect to the state acquired by the acquisition means,
The prediction means predicts when the abnormality occurs in the monitoring target device when the distance to the abnormal space calculated by the calculation means is shorter than the distance to the normal space calculated by the calculation means. Do,
The prediction apparatus according to claim 2, wherein: A monitoring item other than the monitoring item related to the abnormality of the prediction target, and the change in the state detected by some of the monitoring target devices in which the abnormality has occurred is common to the standard or more An additional means for adding a monitoring item having a property to a monitoring item related to the abnormality of the monitoring target device having a device attribute that identifies the part of the monitoring target devices;
The prediction device according to any one of claims 1 to 3, wherein the prediction device is characterized in that: A monitoring target device and a prediction device,
The monitoring target device includes a detection unit that detects a state of a monitoring item related to the abnormality of the prediction target,
The prediction device includes an acquisition unit that acquires a status of a monitoring item related to the abnormality detected by the monitoring target device from the monitoring target device, and the abnormality has occurred with respect to the state acquired by the acquisition unit. A calculation unit that calculates a distance to an abnormal space that is predetermined as a space that represents a state in the case of being present, and the occurrence of the abnormality in the monitoring target device is predicted based on the distance to the abnormal space that is calculated by the calculation unit A prediction means for
A prediction system characterized by that. On the computer,
An acquisition function for acquiring a state detected by a monitoring target device for a monitoring item related to a prediction target abnormality;
For the state acquired by the acquisition function, a calculation function that calculates a distance to an abnormal space that is predetermined as a space that represents the state when the abnormality occurs;
A prediction function for predicting the occurrence of the abnormality in the monitoring target device based on the distance to the abnormality space calculated by the calculation function;
A program to realize

Images