JP2013029987A - Monitor system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for precisely making a maintenance determination as a technique for determining whether maintenance is necessary on the basis of the trend of occurrence of an event in an image formation device which was requested to be maintained in the past.SOLUTION: On the basis of maintenance information stored in a maintenance information storage unit 64 and event information stored in an event information storage unit 16 with respect to an image formation device 10 for which maintenance operation was carried out in the past in response to an emergency call etc., a reference occurrence pattern generation unit 65 extracts an occurrence trend (occurrence pattern) of events in time series with respect to a period (a period defined by tracking back maintenance operation to a start point) of predetermined length before the maintenance operation, specifies a combination of types (event codes) of events occurring with correlation (co-occurring), and generates and stores an occurrence pattern (reference occurrence pattern) on which maintenance is determined in a reference occurrence pattern storage unit 66 for each event code associated with the combination.

Description

  The present invention relates to a monitoring system and a program.

  For example, in an image forming apparatus having an image forming function for forming and outputting an image on a recording material such as paper, when an event (such as an abnormality such as a paper jam or a transfer failure) that interferes with the operation of the image forming function occurs frequently, This is inconvenient for the user of the image forming apparatus. Therefore, it is desired to perform a quick maintenance operation on an image forming apparatus that has fallen into such a situation.

Until now, various inventions have been proposed regarding techniques for monitoring the state of an apparatus and determining the necessity of maintenance.
For example, an operation recording device that records an operation situation that occurred within a predetermined time difference in an operation status record file together with its occurrence time information, and the operation situation record file is monitored, and a combination of a plurality of operation situations exists in the related database. Then, if the difference between the occurrence time information is within a predetermined time difference, an invention of a maintenance management device having a record monitoring device for predicting and outputting a failure result that may occur in the future stored in a related database to a display device is proposed. (See Patent Document 1).

  For example, in an abnormality processing method that detects an abnormality of an information processing device and selects and executes an abnormality handling process corresponding to the detected abnormality of the information processing device, operation mode information indicating the operation status of the information processing device and information An abnormality management table that correlates and stores an abnormality handling process that corresponds to an abnormality of the processing device, and an abnormality that determines the operation mode information and selects an abnormality handling process that corresponds to the determined operation mode information from the abnormality management table An invention of an abnormality processing system provided with a processing mechanism has been proposed (see Patent Document 2).

  For example, an input unit that collects failure information including a failure type indicating the type of failure that occurred and an occurrence time indicating the time when the failure occurred from the monitored device, and defines the failure type of the failure information to be reported Reporting the failure information to a maintenance center that monitors the monitored device in response to a result determined by the filter unit, a filter unit that determines the failure information to be reported based on the definition table And a registration unit that extracts the failure type to be reported to the maintenance center based on the failure information collected by the input unit, and updates the definition table with the extracted failure type. An invention of a failure notification device that automatically updates the failure type of the failure information to be notified to the maintenance center in accordance with the operation of the failure notification device has been proposed. See Patent Document 3).

JP 2001-331350 A JP 2001-209561 A JP 2004-320267 A

  The present invention relates to a technique for determining whether or not maintenance is necessary based on a tendency of occurrence of an event in an image forming apparatus that has reached a maintenance request in the past, and an object thereof is to propose a technique that enables a maintenance determination to be performed with high accuracy. And

  The present invention according to claim 1 is based on an event code indicating the type of an event that has occurred in the apparatus, collected for each case of maintenance work performed on the apparatus, and a plurality of cases for all maintenance work cases. For each combination of the event codes identified by the identifying means, the identifying means for identifying a combination of the plurality of event codes in which the ratio of the number of maintenance work cases in which the event codes occur is equal to or greater than the first threshold, The setting means for setting the occurrence tendency of the event code included in the combination as a determination criterion for determining the necessity of performing maintenance work on the monitored device, and the combination of the event code specified by the specifying means, Calculating the similarity between the occurrence tendency of the event code generated in the monitored device and the occurrence tendency of the event code set as the determination criterion by the setting means; And a determination unit that determines that the monitored device needs to be subjected to maintenance work when all of the calculated similarities for each event code are equal to or greater than a second threshold. System.

  According to a second aspect of the present invention, in the first aspect of the present invention, the specifying unit, the setting unit, and the determining unit perform processing in a classification unit of maintenance work, and the monitoring system is a case of maintenance work. The monitoring system further comprises subdividing means for subdividing the classification of maintenance work as the number increases.

  According to a third aspect of the present invention, in the first or second aspect of the present invention, the specifying means includes the first event code and the first event code for the number of all maintenance work cases calculated by association analysis. Maintenance in which the first event code and the second event code are generated with respect to the support degree indicating the ratio of the number of maintenance work cases in which the event code 2 is generated and the number of maintenance work cases in which the first event code is generated And a certainty factor representing the ratio of the number of work cases, each of which has a threshold value, and a combination of the first event code and the second event code is calculated by calculating the support level and the certainty factor by association analysis. The monitoring system is characterized in that a combination of these event codes is specified when the degree of support and the certainty are each equal to or greater than the threshold value.

  According to a fourth aspect of the present invention, in the first aspect of the present invention according to the first to third aspects, the specifying unit performs maintenance work on the combination of the specified plurality of event codes, although the plurality of event codes are generated from each other. The ratio of the number of maintenance work cases in which the plurality of event codes have occurred to the number of cases where the event code has not been applied is calculated, and combinations that have the calculated ratio equal to or greater than a reference value are further specified and given to the setting unit This is a monitoring system characterized by that.

  The present invention according to claim 5 is the number of all maintenance work cases based on the event codes indicating the types of events that have occurred in the apparatus, collected for each case of maintenance work performed on the apparatus. A specific function for identifying a combination of the plurality of event codes in which the ratio of the number of maintenance work cases in which a plurality of event codes have occurred is equal to or greater than a first threshold, and for each combination of event codes identified by the specific function A setting function for setting an event code generation tendency included in the combination as a determination criterion for determining the necessity of performing maintenance work on the monitored device, and a combination of the event codes specified by the specific function The occurrence tendency of the event code generated in the monitored device and the occurrence tendency of the event code set as the determination criterion by the setting function. A determination function that calculates similarity and determines that the monitored device needs to be subjected to maintenance work when all of the calculated similarities for each event code are equal to or greater than the second threshold is realized. It is a program for.

  According to the present invention according to claims 1 and 5, it is possible to set a determination criterion capable of performing maintenance determination with higher accuracy than in the case where the present invention is not applied.

  According to the second aspect of the present invention, the determination criteria can be set finely as the number of maintenance work cases increases.

  According to the third aspect of the present invention, it is possible to specify a combination of event codes to be set as determination criteria by using the result of association analysis.

  According to the fourth aspect of the present invention, it is possible to specify a combination of event codes estimated to be highly likely to lead to maintenance work, and set it as a determination criterion.

It is a figure which illustrates the structure of the monitoring system which concerns on one Embodiment of this invention. 1 is a diagram illustrating a structure of an image forming apparatus. It is a figure which illustrates the composition of a standard generation pattern generation part. It is a figure which illustrates the data structure of maintenance information. It is a figure which illustrates the mode of processing by a maintenance information acquisition part. It is a figure which illustrates the processing flow by a maintenance information acquisition part. It is a figure which illustrates the processing flow by a code classification determination part. It is a figure which illustrates the code classification and subdivision level according to the number of maintenance work. It is a figure which illustrates the processing flow by a maintenance classification | category classification | category part. It is a figure which illustrates the mode of classification (grouping) by a maintenance classification part. It is a figure which illustrates the processing flow by a merge part. It is a figure which illustrates the combination (merge) result by a merge part. It is a figure which illustrates the result of having performed association analysis. It is a figure which illustrates the processing flow by a relevance analysis part and a reference | standard generation pattern setting part. It is a figure which illustrates the processing flow by a similarity calculation part and a maintenance necessity determination part. It is a figure which illustrates the hardware constitutions of a monitoring apparatus.

A monitoring system according to an embodiment of the present invention will be described with reference to the drawings.
First, the background will be described prior to the description of the monitoring system according to the present example.
Conventionally, regarding maintenance of a monitored apparatus (for example, an image forming apparatus) that can be remotely monitored using wired or wireless communication, event information in each monitored apparatus registered in a database (for example, without manual intervention) (for example, There is a monitoring system that acquires and analyzes (Fault log or Fail code) and issues a warning based on the analysis result.

  Here, a Fail code (an example of event information) generated in the image forming apparatus serving as a monitored apparatus is, for example, a notification of detection of an abnormal state within the range of the sensor, and is a part of the abnormal state that the system falls into There are many cases that only show. That is, for example, in an image quality adjustment cycle (calibration) that is intermittently performed in the image forming apparatus, a fail code that is generated when a pattern for adjustment cannot be normally detected is generated by erroneous detection or the like. Sometimes. For this reason, the occurrence of the Fail code does not mean that immediate maintenance is required.

As described above, since the occurrence of a fail code is not always in a state that requires maintenance, not only the occurrence of a fail code is detected, but also the analysis result of the occurrence tendency and the number of times of the noticed fail code. Therefore, it is estimated whether or not the apparatus is in an abnormal state, and the necessity of maintenance work is determined from the estimation result. Furthermore, the accuracy of determining the necessity of maintenance is improved by weighting importance or the like for each fail code of interest to improve the estimation accuracy of the abnormal state.
That is, it is desirable that the status of the monitored device is expressed by each fail code including the time series occurrence state, but in reality, the status of the monitored device can be expressed only by each fail code. Since there is no failure, the occurrence tendency and frequency of the fail code, and further weighting for each fail code, it is estimated from the corresponding fail code that the maintenance is necessary, and the necessity of maintenance is required. Judgment has been made.

However, as described above, the Fail code output from the monitored device is, for example, a notification of abnormal state detection within the range of the sensor, and only indicates a part of the abnormal state that the system falls into. For this reason, a partial abnormal state is detected in the entire system, but the state of the entire monitored device cannot be expressed.
For example, maintenance may be requested for only a part of a plurality of monitored devices in which the number of occurrences of the target fail code exceeds a threshold value during a certain period. On the other hand, maintenance may be requested even for a monitored apparatus in which the number of occurrences of the target fail code is less than a threshold value. In such a case, it is difficult to accurately determine the necessity of maintenance even if attention is paid only to the target fail code.

It is only necessary that the status of the monitored device can be accurately expressed by the code information. However, there are many types of code information indicating the status of the monitored device (for example, 1000 or more types in the image forming apparatus), and the status that can occur in the monitored device It is not realistic to exhaustively define the data because the amount of data becomes enormous.
Therefore, the monitoring system of this example improves the detection accuracy of monitored devices that require maintenance by adopting the following configuration.

FIG. 1 illustrates the configuration of a monitoring system according to an embodiment of the present invention.
The monitoring system of this example includes an image forming apparatus 10 that forms and outputs an image on a recording material such as paper, and a maintenance information input terminal 50 that is used by an administrator or a maintenance person of the image forming apparatus 10. Have. In the example of FIG. 1, two image forming apparatuses 10 and one maintenance information input terminal 50 are shown, but the number of these is arbitrary.

  The monitoring system of this example is connected to each of the image forming apparatus 10 and the maintenance information input terminal 50 via a communication network so that they can communicate with each other by wire or wirelessly, and the information collected from these is used for the image forming apparatus 10. It has a monitoring device 60 for determining the necessity of maintenance work. In the example of FIG. 1, the monitoring device 60 is configured by a single device, but each functional unit may be distributed to a plurality of devices.

  The image forming apparatus 10 is an apparatus having an image forming function for forming and outputting an image on a recording material such as paper. Examples of the image forming apparatus include apparatuses such as a printer (document printing apparatus), a copier (document copying apparatus), a facsimile apparatus (document transfer apparatus), and a multi-function apparatus having a combination of functions of these apparatuses. included.

The operation of the image forming apparatus 10 will be schematically described with reference to FIG.
FIG. 2 illustrates the structure of the image forming unit in the image forming apparatus 10.
The image forming apparatus 10 of this example is an intermediate transfer method generally called a tandem type, and as a representative functional unit, a plurality of image forming units 1Y, 1M, 1C on which toner images of respective color components are formed by an electrophotographic method. , 1K, and the respective color component toner images formed by the image forming units 1Y, 1M, 1C, and 1K are sequentially transferred (primary transfer) to the intermediate transfer belt 15, and transferred onto the intermediate transfer belt 15. A secondary transfer unit 20 that collectively transfers (secondary transfer) the superimposed toner image onto paper P (an example of a recording material), and a fixing unit 34 that fixes the secondary transferred image onto the paper P. Yes.
The image forming apparatus 10 of the present example also includes a control unit 40 that controls the operation of each unit, and a user interface (UI) 41 for receiving information presented to the user and instructions from the user.

In this example, each of the image forming units 1Y, 1M, 1C, and 1K includes a photosensitive drum 11 (11Y, 11M, 11C, and 11K) that rotates in the arrow A direction. Further, around each of the photosensitive drums 11, a charger 12 that charges the photosensitive drum 11, an exposure unit 13 that writes an exposure beam Bm on the photosensitive drum 11 to write an electrostatic latent image, and each color component toner And a developing device 14 that visualizes the electrostatic latent image on the photosensitive drum 11 with toner, and each color component toner image formed on the photosensitive drum 11 is transferred to the intermediate transfer belt 15 by the primary transfer unit 10. Various types of electrophotographic devices such as a primary transfer roll 16 for transferring to the drum and a drum cleaner 17 for removing residual toner on the photosensitive drum 11 are sequentially arranged.
These image forming units 1Y, 1M, 1C, and 1K are arranged substantially linearly in the order of yellow (Y), magenta (M), cyan (C), and black (K) from the upstream side of the intermediate transfer belt 15. The intermediate transfer belt 15 can be contacted and separated.

  In addition, the image forming apparatus 10 of the present example includes a sheet feeding mechanism unit 31 that performs a sheet feeding operation of taking out the sheet P from the sheet storage unit and feeding it to the secondary transfer unit 20 as a sheet transport system, and the secondary transfer unit 20. The conveying belt 32 that conveys the paper P that has passed through the fixing device 34, the fixing inlet guide 33 that guides the paper P to the inlet of the fixing device 34, and the paper P discharged from the fixing device 34. A paper discharge guide 35 and a paper discharge roll 36 for discharging the paper P guided by the paper discharge guide 35 to the outside of the apparatus are provided.

  That is, the paper P fed from the paper storage unit to the secondary transfer unit 20 by the paper feed mechanism unit 31 is electrostatically transferred to the toner image on the intermediate transfer belt 15 by the secondary transfer unit 20. The paper is transported to the transport belt 32 while being peeled from the intermediate transfer belt 15. Then, the toner is conveyed by the conveyance belt 32 to the fixing device 34 through the fixing inlet guide 33 in accordance with the operation speed of the fixing device 34. The unfixed toner image on the paper P conveyed to the fixing device 34 is fixed on the paper P by receiving a fixing process in which heat and pressure are applied by the fixing device 34. Thereafter, the paper P on which the fixed image is formed is conveyed to a paper discharge container (not shown) provided outside the apparatus via a paper discharge guide 35 and a paper discharge roll 36.

Further, the image forming apparatus 10 of this example has a function of generating event information by detecting various events occurring in the apparatus during the image forming operation. The event information mainly includes general code information related to abnormal operation. In this example, a predetermined event is detected as a detection target, and events such as warning (warning) and information (information) are detected in addition to errors (failures).
Each event information is identified by an event code (in this example, a machine internal generated code) assigned in advance for each event type. For example, “75-XXX” is assigned to the event information when the paper feed from the manual paper feed unit is unsuccessful, and “127-YYY” is assigned to the event information when the punch waste container is nearly full. Code is assigned. In this example, 400 or more types of codes are assigned only to the image forming unit.
For example, when there is a visit request (maintenance request) that “a jam occurs in the fuser unit (fixing device 34) and the sheet is not discharged”, an event that actually occurs in the image forming apparatus 10 is a fixing device. 34 or a running failure due to wear of the paper transporting roll of the input / output unit of the fixing device 34. That is, a plurality of pieces of event information having different types (event codes) can be generated from one factor.

The event information generated by the image forming apparatus 10 is transmitted to the monitoring apparatus 60 by wireless or wired communication. The event information may be transmitted immediately in response to the generation of the event information. The generated event information is stored in a memory inside the device, and stored when the predetermined transmission condition is satisfied. You may employ | adopt the structure which transmits the event information prepared previously. Specifically, for example, it is transmitted when the operation mode of the device is switched (for example, when switching from the normal mode to the sleep mode), transmitted in response to a request from the monitoring device 60, or a predetermined time. You may transmit at intervals (for example, every day).
In this example, the event information transmitted to the monitoring device 60 includes device ID information (SerialNo in this example) for identifying the own device, an event code indicating the type of the event that occurred (in this example, a machine internal generated code), Includes the date and time of the event.

Next, the maintenance information input terminal 50 will be described.
The maintenance information input terminal 50 of the present example visits the installation location of the image forming apparatus 10 and receives maintenance information related to the maintenance work from a maintenance person (or a person who received the report) who actually performed the maintenance work. The input is received and transmitted to the monitoring device 60. In addition, the maintenance information input terminal 50 of this example receives information about the determination result relating to the necessity of maintenance for the image forming apparatus 10 from the monitoring apparatus 60 and displays the information on a display device provided in the image forming apparatus 10. .
In this example, the maintenance information transmitted to the monitoring device 60 includes device ID information (Serial No in this example) for identifying the image forming apparatus 10 that is the subject of the maintenance work, and the date and time of execution of the maintenance work (in this example, Visit date), information indicating the contents of the maintenance work (in this example, a phenomenon code, a cause code, a work code, and a part code) are included.

Next, the monitoring device 60 will be described.
The monitoring device 60 of this example is a device that determines the necessity of maintenance for the image forming apparatus 10, and includes an event information acquisition unit 61, an event information storage unit 62, a maintenance information acquisition unit 63, a maintenance information storage unit 64, and a reference occurrence. A pattern generation unit 65, a reference generation pattern storage unit 66, a similarity calculation unit 67, and a maintenance necessity determination unit 68 are included.

The event information acquisition unit 61 acquires (receives) event information transmitted from the image forming apparatus 10 and stores it in the event information storage unit 62.
The event information stored (stored) in the event information storage unit 62 includes device ID information for identifying the image forming apparatus 10 (SerialNo in this example), and an event code indicating the type of event that has occurred (in this example, machine Internally generated code), event occurrence date and time, etc. are stored.

The maintenance information acquisition unit 63 acquires (receives) maintenance information transmitted from the maintenance information input terminal 50 and stores it in the maintenance information storage unit 62.
The maintenance information stored (stored) in the maintenance information storage unit 64 includes device ID information (Serial No in this example) for identifying the image forming apparatus 10 subjected to the maintenance work, and the date and time of execution of the maintenance work (this book). In the example, visit date), information indicating the contents of the maintenance work (in this example, a phenomenon code, a cause code, a work code, and a part code) are stored.

  The reference generation pattern generation unit 65 is stored in the maintenance information and event information storage unit 16 stored in the maintenance information storage unit 64 for the image forming apparatus 10 that has been subjected to maintenance work in response to an emergency call or the like in the past. Based on the event information, the trend of occurrence of time-series events (occurrence pattern) is extracted for a period of a predetermined length before the maintenance work (period going back to the maintenance work) and related. A combination of event codes generated (co-occurring) is identified, and an occurrence pattern (reference occurrence pattern) serving as a reference for maintenance determination is generated for each event code related to the combination and stored in the reference occurrence pattern storage unit 66 .

The reference generation pattern generation unit 65 will be described in detail.
The reference generation pattern generation unit 65 of this example includes a maintenance information acquisition unit 71, a code type determination unit 72, a maintenance type classification unit 73, a maintenance type classification result storage unit 74, event information, as illustrated in FIG. A forming unit 75, a merge unit 76, a relevance analysis unit 77, and a reference generation pattern setting unit 77 are included.

The maintenance information acquisition unit 71 acquires the maintenance information stored in the maintenance information storage unit 64.
FIG. 4 illustrates the data structure of maintenance information used in this example.
As shown in FIG. 4A, the maintenance information in this example includes a “model code” indicating the model of the image forming apparatus 10 subjected to maintenance work, and a “example” that identifies a case (maintenance history) for each model. “No”, “SerialNo” (device ID information) for identifying the image forming apparatus 10 in units of devices, “visit date” indicating the date and time of visit for maintenance work (implementation date and time of maintenance work), failure location / failure phenomenon / It has a “phenomenon code” indicating the state, a “cause code” indicating the cause of the failure, and a “work code” indicating the action content (and the part replaced when the replacement occurs).
FIG. 4B shows an example of setting a phenomenon code. In this example, a 1-digit code indicating a failure location, a 1-digit code indicating a failure phenomenon, and a 3-digit code indicating a state are shown. It is a specification to set the 5-digit code.
FIG. 4C shows an example of setting the cause code. In this example, the specification is such that a two-digit code indicating the cause of the failure is set.
FIG. 4D shows an example of setting the work code. In this example, the specification is such that a two-digit code (and a part number for identifying the replaced part) indicating the treatment content is set.

Since the reference generation pattern generation unit 65 of this example generates a reference generation pattern for each model of the image forming apparatus 10, the maintenance information acquisition unit 71 extracts maintenance information about the target model from all the maintenance information. Thus, numbering is performed for each model (numbers (case numbers) are given in chronological order).
FIG. 5 illustrates a state of processing by the maintenance information acquisition unit 71. FIG. 5A shows an example of maintenance information stored in the maintenance information storage unit 64. FIG. 5B shows an example in which maintenance information of model code = “AB” is extracted from the maintenance information and numbered. It is.

FIG. 6 illustrates a processing flow by the maintenance information acquisition unit 71.
That is, the maintenance information acquisition unit 71 of this example refers to the maintenance information storage unit 64, checks whether there is maintenance information that has not yet been classified by model (step S11), and has unprocessed maintenance information. The maintenance information is classified by model based on the model code (step S12). This is repeated until there is no unprocessed maintenance information.

The code type determination unit 72 determines a code type as a reference for classifying (grouping) maintenance information for the maintenance information of the model type acquired by the maintenance information acquisition unit 71.
In this example, as illustrated in the processing flow of the code type determination unit 72 in FIG. 7, the code type that serves as a reference for classifying the maintenance information for each model according to the case number (number of maintenance work cases) counted for each model. Is determined (step S21).

Specifically, for example, in a state where 5000 pieces of maintenance information of the target model are not accumulated, it is determined that classification is not performed based on only a phenomenon code in a state where 5000 pieces are not accumulated, In the state where 12,000 records are accumulated, the classification is subdivided as the number of maintenance work cases increases, for example, the classification is determined based on the phenomenon code and the part number (and the work code).
The number of maintenance work cases that serve as the standard for determining the classification level (subdivision level) is, for example, the number of cases for 3 months, 6 months, and 1 year according to the maintenance occurrence rate predicted when the monitored device is introduced to the market. Set based on.
In the initial state, the number of maintenance information items to be classified (the number of maintenance work cases) is small, and the number of cases increases with the passage of time. In 10), a state requiring maintenance is widely recognized, and classification (subdivision) is performed using detailed information as the number of cases increases.

FIG. 8 illustrates code types and subdivision levels according to the number of maintenance operations.
In the example of FIG. 8, when the number of cases is less than 3000, the subdivision is not performed, and when the number of cases is 3000 or more and less than 10,000, the first subdivision (subdivision level 1) is performed based on the phenomenon code. ), If the number of cases is 10000 or more and less than 20000, perform the second stage of subdivision (subdivision level 2) based on the phenomenon code and part number (and work code), and the number of cases is 20000 In the above case, the third subdivision (subdivision level 3) is performed based on the phenomenon code, the part number (and work code), and the cause code.

The maintenance type classification unit 73 sequentially reads out the maintenance information of the model type acquired by the maintenance information acquisition unit 71 and classifies it according to the subdivision level determined by the code type determination unit 72.
FIG. 9 illustrates a processing flow by the maintenance type classification unit 73.
That is, the maintenance type classification unit 73 of this example confirms the code type (code type serving as a reference for classification) determined by the code type determination unit 72 (step S31), and there is no unprocessed (unclassified) data. The following processing is repeated until (step S32).
First, the phenomenon code, component code, and cause code are classified according to the code type determined by the code type determination unit 72 (code type as a reference for classification) (step S33). Next, a set (case group) in which cases of the same phenomenon code, part code, and cause code are collected is created (step S34). Thereafter, it is determined whether the number of configuration data (number of cases) is equal to or greater than a predetermined threshold (5 in this example) (step S35), and a set whose number of cases is equal to or greater than the threshold is stored as a maintenance type classification result. The information is stored in the unit 73 (step S36).
Note that in this example, a group whose number of cases is less than the threshold is not classified at the subdivided level, but is only classified at the previous level.

FIG. 10 illustrates an example of classification (grouping) by the maintenance type classification unit 73.
Describing according to the example of FIG. 8, when the number of cases is 3000 or more and less than 10,000, classification using only a phenomenon code (subdivision level 1) is applied. As a result, for example, the classification A that collects maintenance information of the phenomenon code = “75100” is extracted, and similarly, the classifications B, C, and D that collect other maintenance information for each phenomenon code are extracted.
Further, when the number of cases is 10000 or more and less than 20000, classification (subdivision level 2) using a phenomenon code and a part number (and work code) is applied. As a result, for example, the classification A-1 that collects the maintenance information of the part code = “74870B22” that is the number of cases equal to or greater than the threshold in the classification A is extracted, and similarly, the other classifications B, C, and D are subdivided. B-1, B-2, B-3, etc. are extracted.
Further, when the number of cases is 20000 or more, classification using a cause code (subdivision level 3) is applied. As a result, for example, the classification A-1-1 in which the maintenance information of the cause code = “30”, which is the number of cases equal to or greater than the threshold value in the classification A-1, is extracted, and similarly, the other classifications B-1, B B-1-1, B-2-1, B-3-1 and the like obtained by subdividing -2 and B-3 are extracted.
Note that “None” in the column of category 2 (subdivision level 2) and category 3 (subdivision level 3) in FIG. 10 indicates that the number of cases is less than the threshold. In this case, category 1 (subdivision) Level 1) is applied.

Based on the maintenance information given from the merge unit 76, the event information forming unit 75 reads event information related to the maintenance information from the event information storage unit 62 and forms it into a format suitable for the subsequent processing.
In this example, based on the maintenance information given from the merge unit 76, it is event information about the image forming apparatus 10 (the image forming apparatus 10 with the same Serial No.) that is the object of the maintenance work, and is the visit date. (Maintenance work date / time) Collects event information of all types (event codes) in a period of a predetermined length before (for example, the visit date and the previous three days), and for each event code, Generate molding data that lists the number of occurrences by day. Note that event codes whose total number of occurrences is less than a threshold (for example, 5) are unlikely to have led to maintenance work. By excluding them, the burden of subsequent processing is reduced. .

  The merge unit 76 refers to the classification result stored in the maintenance type classification result storage unit 73, and for each classification, the event code formed by the event information forming unit 75 for each case (maintenance information) included in the classification. Each piece of molding data (data in which the number of occurrences of events is arranged for each day) is acquired, and combined data is generated by combining (merging) the maintenance information with the respective molding data related thereto.

FIG. 11 illustrates a processing flow by the merge unit 76.
That is, the merging unit 76 of this example repeats the following processing for each classification until there is no unprocessed (uncoupled) data (step S41).
First, by providing each event (maintenance information) included in the target classification to the event information forming unit 75, all types (from the inside of the machine) from three days before the visit date based on the visit date of the maintenance work and Serial No. Molding data for each event code obtained from the event information of (occurrence code) is acquired (step S42). Next, combined data obtained by combining (merging) the acquired molding data with maintenance information is generated, and the combined data obtained as a result of the merge is supplied to the relevance analysis unit 76 (step S43).

FIG. 12 illustrates the result of merging (merging) by the merging unit 76.
FIG. 12A relates to a plurality of pieces of maintenance information (maintenance information of phenomenon code = “75100” (FEED portion JAM), part code = “74870B22”) classified into the classification A-1 and each maintenance information. It is an example of the joint data which combined the molding data for every event code. FIG. 12B is a graph showing the transition of the number of occurrences of events with the passage of time and date expressed based on molding data (data in which the number of occurrences of events is arranged by day) related to a certain maintenance work example. It is.

  The relevance analysis unit 77 has relevance based on the combination data of the maintenance information supplied from the merge unit 76 and the related molding data (data in which the number of occurrences of events is arranged by day). The combination of event codes that occurred (co-occurred) is extracted. In this example, the relationship between event codes is analyzed by cluster analysis (for example, association analysis), and maintenance work cases in which two types of event codes occur with respect to the number of maintenance work cases of the same classification. A combination of event codes whose number ratio is equal to or greater than a threshold is identified and extracted.

FIG. 13 illustrates the result of performing association analysis on the data example of FIG. In FIG. 13, the degree of support represents the ratio of transactions including item set X (a set of machine generated codes) and item set Y (a set of machine generated codes) to the whole. The certainty factor represents a value obtained by dividing the number of transactions σ (X∪Y) including the item set X and the item set Y by the number of transactions σ (X) including the item set X.
In other words, the support level is the ratio of the number of maintenance work cases in which the event code X and the event code Y have occurred to the number of cases in all maintenance work (the case in which the event code X and the event code Y have occurred is all cases) Ratio). The certainty factor is the ratio of the number of maintenance work cases in which event code X and event code Y have occurred to the number of maintenance work cases in which event code X has occurred (event code Y occurs when event code X occurs) Of the cases that have been completed).
According to the example of FIG. 13, among the combinations having a certainty factor of 0.5 or more, in the classification A-1 (maintenance information of phenomenon code = “75100” (FEED part JAM), part code = “74870B22”), The support level of the machine internal generation code = “75-100” is high, and in the correlation rule, the support levels of the machine internal generation code = “75-100” and “07-135” are high.

  In the relevance analysis unit 77 of this example, a combination of two machine internal generated codes (event codes) in which the certainty and support of the correlation rule obtained by association analysis are equal to or greater than a threshold (0.5 in this example). According to the example of FIG. 13, a combination of “75-100” and “07-135” and a combination of “75-100” and “42-106” are extracted. That is, a combination having a high certainty factor and a high support factor is specified and extracted. In this example, a combination of “()” and “75-100” (that is, “75-100” alone) is also extracted. In addition, if it is set as the structure which calculates a certainty factor and a support degree for every combination of 3 or more event codes, the combination of 3 or more event codes which have relevance can also be extracted. In addition, although the same value is used for the threshold value of confidence and the threshold value of support in this example, different values may be used.

  The reference occurrence pattern setting unit 78 uses the combination of event codes extracted by the relevance analysis unit 77 as a candidate to determine whether to generate a reference occurrence pattern that is a determination criterion for determining the necessity of performing maintenance work. The reference generation pattern is generated for the corresponding combination and stored in the reference generation pattern storage unit 66.

  For example, in the classification A-1 (maintenance information of phenomenon code = “75100” (FEED part JAM), part code = “74870B22”), the machine generated codes (event codes) of “75-100” and “07-135” However, if this occurs even in a steady state, the necessity for the maintenance work is erroneously determined. Therefore, the number of generated codes generated in the machine of “75-100” and “07-135” is calculated for the maintenance work and the steady state (when the maintenance work is not performed). The ratio between the number of occurrences (number of occurrences when maintenance work is performed) and the number of steady occurrences (number of occurrences in a steady state) is equal to or greater than a reference value (that is, the number of occurrences of maintenance / the number of steady occurrences ≧ reference value (for example, 0. 3)), it is determined that the combination of machine internal generated codes that require a high degree of maintenance work is performed, and the generation pattern (time-series generation tendency) of the machine internal generated codes related to the combination is determined as the reference generation pattern. Is set (stored in the reference generation pattern storage unit 66). In this example, an occurrence pattern of a code generated in a machine as a target is extracted for each maintenance work example, and an average of these is used as a reference occurrence pattern.

That is, for example, in the cases of “75-100” and “07-135”, if the number of maintenance occurrences = 3 and the number of normal occurrences = 10, 3/10 ≧ 0.3 (= reference value), A reference generation pattern is generated for the case. Further, for example, in the case of “75-100” alone, if maintenance occurrence count = 3 and steady occurrence count = 112069, 3/12069 <0.3 (= reference value). Does not generate an occurrence pattern. For example, in the cases of “75-100” and “42-106”, if the number of maintenance occurrences = 2 and the number of steady occurrences = 377, 2/377 <0.3 (= reference value). No reference generation pattern is generated for this case.
As described above, whether the combination of the event codes extracted by the relevance analysis unit 77 actually leads to the maintenance work or whether it occurs regardless of the maintenance work is the maintenance occurrence ratio (number of maintenance occurrences / number of steady occurrences). Based on the verification, if the frequency of occurrence other than when maintenance occurs (steady state) is low, it is presumed that there is a high possibility that it will lead to maintenance work, so a reference occurrence pattern is generated for that combination.

FIG. 14 illustrates a processing flow by the relationship analysis unit 77 and the reference generation pattern setting unit 78.
That is, the relevance analysis unit 77 and the reference generation pattern setting unit 78 of this example repeat the following process for each classification until there is no unprocessed data (step S51).
First, the relevance analysis unit 77 performs cluster analysis based on the combined data of the maintenance information supplied from the merge unit 76 and the related molding data (data in which the number of occurrences of events are arranged by day). Then, based on the result of the cluster analysis, a combination of event codes related to each case (maintenance information) included in the target classification is extracted (step S53). Next, the reference occurrence pattern setting unit 78 calculates the maintenance occurrence ratio by calculating the number of maintenance occurrences and the number of steady occurrences for the combination of event codes extracted by the relevance analysis unit 77 (step S54). A reference generation pattern is set by specifying a combination in which is equal to or greater than the reference value (step S55).

  The similarity calculation unit 67 generates an event based on event information acquired from the image forming apparatus 10 in which a new event has occurred and accumulated in the event information storage unit 62 for the combination of event codes for which the reference occurrence pattern is set. The occurrence pattern of the latest event is extracted for each code, and the similarity (for example, correlation value) with the reference occurrence pattern in the event type is calculated.

The maintenance necessity determination unit 68 compares the similarity for each event code calculated by the similarity calculation unit 67 with a predetermined threshold for the combination of event codes for which the reference occurrence pattern is set, and determines all similarities. Is greater than or equal to the threshold value, it is determined that maintenance work needs to be performed on the image forming apparatus 10. Further, by associating the contents of maintenance work with the combination of event codes, it is possible to determine what kind of maintenance work should be performed.
The threshold value may be a fixed value, but the determination accuracy may be improved by using a variable value corresponding to the number of occurrences of event information. That is, for example, when the number is less than 100, 0.5 is used as the threshold, when the number is less than 1000, 0.7 is used, and when the number is 1000 or more, 0.9 is used as the threshold. .

FIG. 15 illustrates a processing flow by the similarity calculation unit 67 and the maintenance necessity determination unit 68.
That is, first, the similarity calculation unit 67 extracts event information within the latest predetermined length period related to the image forming apparatus 10 that is a target for determining the necessity of maintenance from the event information storage unit 62. Processing is performed (step S61), and it is determined whether or not the corresponding event information exists (step S62).
If the corresponding event information does not exist, the process is terminated. On the other hand, if the corresponding event information exists, the combination of event codes for which the reference occurrence pattern is set, the latest event for each event type The occurrence pattern is extracted (step S63), and the similarity with the reference occurrence pattern in the event code is calculated (step S64). Then, it is determined whether the similarity in all event codes related to the combination is greater than or equal to a threshold (step S65), and if the similarity in all event codes related to the combination is greater than or equal to the threshold, It is determined that maintenance work needs to be performed on the image forming apparatus 10 (step S66).

  As described above, in the monitoring system of the present example, a combination of event codes that are generated with relevance (co-occurred) in a predetermined period before maintenance work is specified, and each event code related to the combination is identified. In this configuration, a reference generation pattern is generated to determine the necessity of maintenance work, whereby the determination accuracy of the necessity of maintenance work can be improved.

  In the above example, the maintenance information is classified, the maintenance work that has been classified (grouped) is extracted, event information (molding data) for the maintenance work is acquired, and merged with the maintenance work. doing. As a result, the amount of data to be processed is reduced, and the processing speed can be increased. However, when re-creating the reference generation pattern, it is necessary to perform the above-described process again. Therefore, it is possible to speed up the re-creation of the reference generation pattern by acquiring event information (molded data) at the time of maintenance information acquisition, combining (merging), and then performing classification. .

In the above example, a determination criterion setting device that sets a determination criterion for whether maintenance is necessary by specifying a combination of event codes, and determines the necessity of maintenance for the image forming apparatus 10 based on the set determination criterion. Although the determination device and the monitoring device 60 are integrally provided, these may be configured as separate devices.
Further, for example, the monitoring device 60 may be operated as a determination criterion setting device, and each of the image forming apparatuses 10 may be operated as a determination device. That is, the monitoring device 60 transmits the created determination criterion to each image forming device 10, and each image forming device 10 determines whether or not maintenance is required for itself based on the determination criterion received from the monitoring device 60. Configure.
Further, the image forming apparatus 10 related to the setting of the reference generation pattern and the image forming apparatus 10 to be subjected to maintenance determination may be the same apparatus or different apparatuses. That is, by setting a reference generation pattern based on past event information and maintenance information in one or more image forming apparatuses 10 (which may include the image forming apparatus 10 to be subjected to maintenance determination), the reference The necessity of maintenance for the image forming apparatus 10 subject to maintenance determination can be determined using the generated pattern.
Further, an apparatus other than the image forming apparatus 10 may be set as a monitoring target (monitored apparatus), and the necessity of maintenance can be determined by monitoring the states of various apparatuses.

FIG. 16 illustrates the hardware configuration of the monitoring device 60.
In this example, a main memory such as a CPU (Central Processing Unit) 81 that performs various arithmetic processing, a RAM (Random Access Memory) 82 that is a work area of the CPU 81, and a ROM (Read Only Memory) 83 that records basic control programs. Device, auxiliary storage device such as HDD (Hard Disk Drive) 84 for storing programs and various data according to an embodiment of the present invention, display device for displaying and outputting various information, and operations used for input operations by the operator Hardware resources such as an input / output I / F 85 that is an interface with an input device such as a button or a touch panel, and a communication I / F 86 that is an interface for performing wired or wireless communication with other devices are included in the monitoring device 60. The computer has.
Then, the program according to the embodiment of the present invention is read from the auxiliary storage device 84 and the like, loaded into the RAM 82, and executed by the CPU 81, thereby realizing the above-described functional units on the computer of the monitoring device 60. Yes.

Note that the program according to the embodiment of the present invention is based on, for example, the server according to the present 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 network. Set in the computer of the device.
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.

10: Image forming device 50: Maintenance information input terminal 60: Monitoring device
61: Event information acquisition unit, 62: Event information storage unit, 63: Maintenance information acquisition unit, 64: Maintenance information storage unit, 65: Reference generation pattern generation unit, 66: Reference generation pattern storage unit, 67: Similarity calculation unit 68: Maintenance necessity determination unit,
71: Maintenance information acquisition unit, 72: Code type determination unit, 73: Maintenance type classification unit, 74: Maintenance type classification result storage unit, 75: Event information shaping unit, 76: Merge unit, 77: Relevance analysis unit, 77 Reference generation pattern setting section

Claims (5)

Maintenance work in which multiple event codes occur for each number of maintenance work cases based on the event codes indicating the types of events that occurred in the equipment, collected for each maintenance work case performed on the equipment A specifying means for specifying a combination of the plurality of event codes in which the ratio of the number of cases is equal to or more than a first threshold;
For each combination of event codes specified by the specifying means, setting means for setting the tendency of occurrence of the event codes included in the combination as a determination criterion for determining the necessity of performing maintenance work on the monitored device; ,
For the combination of event codes specified by the specifying means, calculate the similarity between the occurrence tendency of the event code generated in the monitored device and the occurrence tendency of the event code set as the determination criterion by the setting means, A determination unit that determines that the monitored device needs to be subjected to maintenance work when all the calculated similarity for each event code is equal to or greater than a second threshold;
A monitoring system characterized by comprising: The specifying unit, the setting unit, and the determination unit perform processing in a classification unit of maintenance work,
The monitoring system further includes subdividing means for subdividing the classification of maintenance work as the number of maintenance work cases increases.
The monitoring system according to claim 1. The specifying means includes a support degree that represents a ratio of the number of maintenance work cases in which the first event code and the second event code are generated to the number of all maintenance work cases calculated by association analysis; And a certainty factor representing the ratio of the number of maintenance work cases in which the first event code and the second event code have occurred to the number of maintenance work cases in which the event code has occurred. When the support level and the certainty factor are calculated by association analysis for the combination of the first event code and the second event code, and the calculated support level and the certainty factor are each equal to or greater than the threshold value, the combination of these event codes Identify
The monitoring system according to claim 1 or 2, characterized by the above. The specifying means, for the combination of the specified plurality of event codes, for the number of cases in which the plurality of event codes are generated each other but no maintenance operation is performed, the maintenance work examples in which the plurality of event codes are generated in each other Calculating a ratio of numbers, further specifying a combination in which the calculated ratio is equal to or greater than a reference value, and giving the combination to the setting unit;
The monitoring system according to any one of claims 1 to 3, wherein: On the computer,
Maintenance work in which multiple event codes occur for each number of maintenance work cases based on the event codes indicating the types of events that occurred in the equipment, collected for each maintenance work case performed on the equipment A specific function for identifying a combination of the plurality of event codes in which the ratio of the number of cases is equal to or greater than a first threshold;
For each event code combination specified by the specific function, a setting function that sets an event code generation tendency included in the combination as a determination criterion for determining the necessity of performing maintenance work on the monitored device; ,
For the combination of event codes specified by the specific function, calculate the similarity between the occurrence tendency of the event code generated in the monitored device and the occurrence tendency of the event code set as a determination criterion by the setting function, A determination function that determines that there is a need to perform maintenance work on the monitored device when all the calculated similarity for each event code is equal to or greater than the second threshold;
A program to realize

Images