JP2008310582A - Maintenance work support apparatus and system, and maintenance work support method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable even an unskilled maintenance worker having little knowledge of a device to perform appropriate restoration work, in restoration work against failure in the device including hardware and software. <P>SOLUTION: An apparatus 1100 is connected to a maintenance object device for monitoring an operation state until the occurrence of failure and outputting it as an operation result file to present support information. The apparatus also acquires a failure signal and a file, transmits/receives them to/from a terminal 1001 to acquire file browsing operation contents for a maintenance worker, counts a display frequency, the presence/absence of input of information, a character string retrieval matching frequency, a frequency of output to the outside, an output frequency of difference information on comparison operation for another file for each kind and each line of the file, creates high frequency information, abstracts important points in the file from the high browsing information to create important point information, acquires restoration work contents, records and stores a work history, acquires a new failure signal and a file from the object device, detects a maintenance work history the same as the new failure signal, calculates similarity between the new file and the detected file based on the important point information, and selects the restoration work contents corresponding to a high similarity file from the maintenance work history to present them. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a maintenance work support apparatus and system and a maintenance work support method, and more particularly to a maintenance method for industrial and consumer devices including a plurality of software and hardware.

Conventionally, the maintenance work of the apparatus is generally performed by experience and knowledge by skilled maintenance personnel.
A normal maintenance operation when a failure occurs includes a cause analysis operation for discovering details of the failure cause and a recovery operation for removing the analyzed cause of failure.
In the cause analysis work, the skilled maintenance staff checks the failure signal such as an alarm or warning output from the device, and browses the operation history information output from the device to identify the detailed cause of the failure. In the case of a device composed of a plurality of hardware and software, the operation history is often composed of a plurality of files instead of a single one, but which file is browsed from the file group and which location in the file is Whether the failure cause analysis is performed by paying attention to is determined by the experience of maintenance personnel and the knowledge of the maintenance target device.
After identifying the cause of the failure, the skilled maintenance staff performs recovery work to remove the cause of the failure. The restoration work can take a plurality of work procedures for one cause of failure. Which work procedure is selected is determined by the experience of the maintenance staff and the knowledge about the maintenance target device.

As described above, the maintenance work of the device depends largely on the experience and knowledge of the maintenance staff for both cause analysis work and recovery work, and it is difficult for an unskilled maintenance staff with little experience and poor knowledge of the maintenance target equipment to perform. It was. As a means for solving such a problem, there is a method of recording and accumulating maintenance work histories by skilled maintenance personnel, and creating and providing support information for unskilled maintenance personnel based on the maintenance work history. (For example, refer to Patent Document 1).
In such conventional technology, when an unskilled maintenance person performs maintenance, the past maintenance work contents of the skilled maintenance staff are presented, and instead of planning an appropriate maintenance work from the beginning, selection is made from the past work contents. By doing so, maintenance work can be performed and the load on unskilled maintenance personnel is reduced.
In particular, in Patent Document 1, maintenance work previously taken by skilled maintenance personnel when a failure occurs in an apparatus is recorded and stored in association with a failure signal of the apparatus. When a non-skilled maintenance person inputs a current failure signal when performing maintenance work, all maintenance work of the skilled maintenance person corresponding to the failure signal is selected and presented to the unskilled maintenance person. An unskilled worker can confirm all the maintenance operations in time series and select a maintenance operation to be performed for the current failure.
JP 2005-122383 A

In the prior art, it is assumed that the maintenance work that can be taken for a specific failure signal is one or a small number of maintenance work, and maintenance is performed by checking all the work of skilled maintenance personnel on the failure signal. It is said that it is possible to learn the work.
However, in a complex device composed of multiple hardware and software, the cause of failure is not always the same for the same failure signal, and different failure causes become the same failure signal. May be notified. In such a case, the contents of the maintenance work differ depending on the cause of the failure.To apply the conventional technology and perform maintenance work by unskilled maintenance personnel, check all maintenance work histories corresponding to the same failure signal as the current failure signal. One maintenance work suitable for the current failure situation is selected from a plurality of types of maintenance work recorded, and the maintenance work is performed.
In this case, since unskilled maintenance personnel have little knowledge of the maintenance target device, only one maintenance operation suitable for the failure status of the current device is selected from a number of maintenance operations corresponding to the current failure signal. There is a problem that is difficult.

  An object of the present invention is to solve the above-mentioned conventional problems, and in the recovery work when a failure of a device including hardware and software occurs, even an unskilled maintenance person who has little knowledge of the maintenance target device can perform appropriate recovery work. A maintenance work support apparatus and system that can be performed, and a maintenance work support method.

  The present invention has a plurality of hardware and software, and is connected to a maintenance target device that monitors the operation status up to the time of the failure and outputs it as an operation result file. When a failure occurs in the maintenance target device A maintenance work support device that presents maintenance support information to maintenance personnel at the same time, acquires a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device, and maintains the acquired operation result file A maintenance staff terminal that is viewed and operated by an operator performs transmission / reception via the user interface, monitors and acquires the browsing operation contents of the operation result file of the maintenance staff on the user interface, and the acquired browsing operation contents Based on the operation result file type, line by line, display frequency on the user interface, information input presence, search by character string search operation At least one of a common frequency, an output frequency to an external output device, and an output frequency of difference information in a comparison operation with another file, and create high-frequency browsing location information with a high count number. The important location information created by identifying and abstracting the important location in the operation result file from the frequency browsing location information is created, the recovery work content of the failure performed by the maintenance staff is acquired, and at least the failure signal and the operation The result file, the important part information, and the maintenance work history including the restoration work contents are recorded and accumulated in correspondence, and when the new failure signal and the new operation result file from the maintenance target device are obtained, the obtained A maintenance work history that is recorded and stored in response to the same failure signal as the new failure signal is detected, and the new location information is based on the important part information in the detected maintenance work history. The similarity between the operation result file and the operation result file in the detected maintenance work history is calculated, and at least one restoration work content corresponding to the calculated operation result file having a high similarity is selected from the maintenance work history. This is a maintenance work support device that selects and presents one.

According to the present invention, when a maintenance work to be performed is selected from maintenance work performed by a skilled maintenance staff in the past, even if there are a large number of maintenance work that can be selected, the current status of the apparatus is maintained in the maintenance work. It is possible to narrow down similar cases.
This makes it easy to select an appropriate maintenance work even when an unskilled maintenance person lacking knowledge of the maintenance target device performs the maintenance work.

The best mode for carrying out the present invention will be described.
According to the maintenance work support method of the present invention, an important point in analyzing a failure is extracted based on a browsing work history of an operation result of an apparatus performed by a skilled maintenance person to analyze the cause of the failure. 2 compares the operation result of the past apparatus with the current operation result, and outputs the recovery work content with the most similar operation result.
When the maintenance target device issues a specific failure signal, an operation result file storing the operation history of the maintenance target device is output to a screen or the like, and browsing and operation operations are recorded in the operation result file by skilled maintenance personnel. Based on the browsing operation frequency information such as the output frequency of the recorded operation result file, the presence / absence of input, the presence / absence of output to an external storage, the output frequency in a search operation, the output frequency in a comparison operation, etc. By extracting locations with high-frequency browsing and operations, it is estimated which locations in the operation result file the skilled maintenance staff focused on for the analysis work. Important parts in the operation result file are recorded as maintenance work history together with maintenance work contents and device information. Here, at least the failure signal, the operation result file, the important location, and the contents of the recovery work performed after failure cause analysis are recorded in the maintenance history record.
After that, when an unskilled maintenance worker performs maintenance work, if the failure signal and the operation result file that are currently generated are input, the past operation result file and the current operation result file for the maintenance work history that matches the failure signal are input. Are compared for important points to calculate the similarity. All maintenance work histories having the same failure signal are compared, and the recovery work contents are extracted and presented to the unskilled maintenance staff for at least one element having a high similarity.
As a result, it is possible to present not only a failure signal but also a restoration operation to a device state similar to the current state from a more detailed device operation result.
Embodiments of a maintenance work support apparatus and system and a maintenance work support method according to the present invention will be described below with reference to the drawings. In addition, this invention is not limited by this Example.

  Example 1 will be described. FIG. 1 is a system configuration diagram of a maintenance target device to which the maintenance work support device according to the first embodiment of the present invention is applied, and FIG. 2 is a software configuration diagram of a maintenance support program in the present embodiment. FIG. 3 is a flowchart illustrating an example of a maintenance history creation procedure in the present embodiment, and FIG. 4 is a flowchart illustrating an example of a restoration work presentation processing procedure. FIG. 6 is a flowchart showing details of the important point abstraction processing (step 3090) shown in FIG. 3, and FIG. 7 shows the abstraction rules related to the browsing point abstraction processing shown in FIGS. It is the figure which showed an example of the important part before and behind abstraction. FIG. 5 is a data structure diagram illustrating an example of a count table of important points according to the present embodiment, and FIG. 8 is a data structure diagram illustrating an example of a maintenance history table according to the present embodiment. FIG. 9 is a flowchart showing an example of a detailed procedure of the comparison process (step 4020) with the apparatus operation log shown in FIG. FIG. 10 is a flowchart showing another example of the count addition process shown in Step 3040 of FIG.

  First, the configuration of the maintenance target apparatus and the maintenance support apparatus in the present embodiment will be described with reference to FIGS. 1 and 2. In FIG. 1, a maintenance target device 1000 receives an operation command input from a user via an operation terminal 1001 or an input device 1004 for the user. The external storage device 1002, the display device 1003, the printing device 1005, and the speaker 1006 provide information such as operation results of the device to the user. In particular, the external storage device 1002 and the printing device 1005 output files stored in the operation result file database 1102 and the maintenance history database 1103, and the display device 1003 and the speaker 1006 are on-line such as failure information in addition to the information in the database. Outputs the operating status. The centralized controller 1100 receives the operation command, reads device information such as device configuration and settings stored in the control information database 1101, and transmits a control command from the communication IF 1104. The operation result file database 1102 records an operation result file group in which the execution status of the maintenance target device 1000 such as control commands transmitted and received is classified and stored by hardware, software, usage, purpose, and the like. The camera controller 1200, the sensor controller 1300, and the actuator controller 1400 receive the control command via the communication IF 1201, the communication IF 1301, and the communication IF 1401, and operate the camera 1202, the sensor 1302, and the actuator 1402 that are the respective control targets. Then, the operation results such as the acquired image, sensor value, and servo position are transmitted to the centralized controller 1100 via the communication IF 1201, the communication IF 1301, and the communication IF 1401.

  In the above configuration, the operation result is stored in the central controller 1100 and the operation result file database 1102 via the communication IF 1201, the communication IF 1301, and the communication IF 1401, and is stored in the external storage device 1002, the display device 1003, the printing device 1005, and the speaker 1006. Display device 1203, external recording 1204, display device 1303, external recording 1304, display device 1403, external recording 1404 connected to the camera controller 1200, sensor controller 1300, and actuator controller 1400. May be output.

  FIG. 2 is a software configuration diagram when a group of programs for realizing the maintenance support method of the present embodiment is stored in the centralized controller 1100. The device operation UI 2001 provides a user interface when the user operates the maintenance target device 1000. The cause analysis UI 2002 provides information necessary for maintenance personnel to perform maintenance work, and displays the contents of the control information database 1101, the operation result file database 1102, and the maintenance history database 1103. In particular, the cause analysis UI 2002 provides information necessary for failure cause analysis to a skilled maintenance staff. The recovery work input UI 2003 accepts and records the work performed by the maintenance staff at the time of failure. The structure of the work content to be accepted may be, for example, a daily work report format in which the work content of a recovery that has occurred at a certain point in time is described in a natural language that other maintenance personnel can understand, or the maintenance target device 1000 or other monitoring system However, the recovery work history data automatically collected by monitoring the work contents of the maintenance staff may be used. The recovery work presentation UI 2004 presents the content of the recovery work extracted from the maintenance history database 1103 by the maintenance support program 2009 to the maintenance staff. The maintenance support setting UI 2005 accepts auxiliary information necessary for processing the maintenance support program 2009. Auxiliary information to be received is, for example, a threshold value or number of extraction when extracting an important part from the output count of the operation result file, an abstraction rule when abstracting the extracted important part, and a presentation when presenting a recovery work It includes at least one of the method and the number of presentations. The failure signal monitoring program 2006 detects a failure by monitoring the communication status and hardware operation status of the maintenance target device 1000 and notifies the user and maintenance personnel of the failure of the maintenance target device 1000 through the device operation UI 2001. The operation result monitoring program 2007 monitors the operation and operation of the maintenance target device 1000 such as the device operation UI 2001 and the control program 2100, and the monitoring content is suitable for each hardware, each software, each monitoring location, each purpose, etc. The operation result file group classified and created in step 1 is recorded in the operation result file database 1102. The maintenance work monitoring program 2008 monitors maintenance work contents performed by maintenance personnel through the cause analysis UI 2002 and other user interfaces. The monitoring result is that the operation result file stored in the operation result file database 1102 is output to any one of the operation terminal 1001, the external storage device 1002, the display device 1003, the printing device 1005, and the speaker 1006 through the cause analysis UI 2002. Includes at least one output operation. When the cause analysis UI 2002 is used, the maintenance support program 2009 cooperates with the maintenance work monitoring program 2008 to identify important parts in the operation result file from the operation history such as the display frequency of the operation result file by the maintenance staff. Extracting important points extracted based on the abstraction rules input from the maintenance support setting UI 2005 and abstracting them as important point features (information), for failure content, important point features, operation result file group, and recovery work presentation The maintenance work history including the recovery work contents input from the UI 2004 is recorded in the maintenance history database 1103. The control program 2100 is software in which the operation of the maintenance target device 1000 is described. Based on the device information stored in the control information database 1101 and the operation command input from the device operation UI 2001, the control program 2100 includes a camera 1202, a sensor 1302, The hardware such as the actuator 1402 is controlled.

In the above description, the cause analysis UI 2002 and the recovery work input UI 2003 are described as separate software, but may be a maintenance work user interface using the same screen.
In addition, although the control program 2100 describes the operation result monitoring program 2007 as another software, an operation result file may be created in the control program 2100 as the same program.
The operation result monitoring program 2007 stores the operation result file in the control information database 1101. However, the operation result monitoring program 2007 can be operated from software such as the cause analysis UI 2002, for example, floppy (registered trademark), compact flash (registered trademark), MMO, It may be stored in an external storage device 1002 such as a CD.

  Next, the operation of each unit in FIGS. 1 and 2 will be described based on the flowcharts in FIGS. The maintenance support program 2009 creates a maintenance work history 8000 according to the processing procedure shown in FIG. 3 when a skilled maintenance worker mainly performs maintenance work using the cause analysis UI 2002, particularly cause analysis work. In addition, the maintenance support program 2009 presents an appropriate recovery work procedure according to the processing procedure shown in FIG. 4 when an unskilled maintenance worker performs a maintenance work mainly using the recovery work presentation UI 2004.

  FIG. 3 shows an example of the processing procedure of the maintenance support program 2009 and related components at the time of maintenance work by skilled maintenance personnel. When a failure signal is reported through the failure signal monitoring program 2007, the cause analysis UI 2002 The procedure for performing maintenance work by analyzing the cause of failure and planning recovery work by the maintenance staff.

  First, when the maintenance staff starts browsing the operation result file using the cause analysis UI 2002, the operation result monitoring program 2008 acquires the failure signal currently notified from the failure signal monitoring program 2007 (step 3010) and displays it. The type of the operation result file output to the apparatus 1003 is determined, and a count table corresponding to the operation result file type is created (step 3020). Further, the part displayed in the cause analysis UI 2002 is monitored in units of lines to obtain the output part of the operation result file (step 3030), and a certain number of count tables corresponding to the currently displayed line are added (steps). 3040). The count process of step 3040 is executed at a constant time interval until browsing of the corresponding operation result file is completed (step 3050). FIG. 5 is a data structure diagram showing an example of the structure of the count table. When the count table 5000 is created in step 3020, the device ID 5010 and the device type 5020 are obtained from the device information held in the maintenance target device 1000. Are recorded, the file name 5030 to be counted, the recording start date and time from the creation date and time of the file, and the total number of lines 5050 of the file are recorded. Further, if necessary, the maintenance staff ID 5060, the maintenance staff name 5070, and the analysis start date 5080 may be recorded from login information of the maintenance staff. Line count information 5100 is created for all lines of the target file, and line number 5110, recording date and time 5120, and description contents 5130 are recorded from the contents of the corresponding file. The browsing count 5140 and the operation count 5150 are numerical data for recording file browsing and operation information by maintenance personnel, and are initialized to 0 when created. In step 3040, by adding the browsing count 5140, the browsing count 5140 corresponding to the row having a long browsing time is increased, and the location in the operation result file analyzed in detail by the maintenance staff can be specified. In the above example, the addition processing to the browsing count 5140 is uniformly added to all lines output through the cause analysis UI 2002. However, the maintenance person pays more attention to the operation result file display method of the cause analysis UI 2002. When the range to be identified can be specified, the added points of the lines displayed in the range may be increased, and the added points of the lines displayed in the other ranges may be decreased. In the text display on a general screen, the central portion of the screen is estimated to be a range that receives more attention, so the added points in the central portion may be increased. In the above example, the browsing count is added according to the display time, but the part that is reviewed by the maintenance staff more than once over the part with a long display time is considered as an important part, and the browsing frequency count is When the count is 0 before the count is started, the added point width is reduced. When the browsing frequency count is larger than 0 before the count is started, the added range may be increased.

  When the maintenance staff starts browsing the new operation result file, a new count table is created (step 3070 returns to step 3020).

  In accordance with the number of extractions set in advance or the threshold value, a row with a large number of counts is extracted as an important part in the count table (step 3080). For example, in the case of extracting an important part by the number of extractions, each row of the count table is sorted in ascending order of the count number, and a line having a higher rank than the extraction number is extracted as an important part. Further, in the case of extracting an important part using a threshold value, a line having a count number equal to or greater than the threshold value is extracted as an important part.

If the maintenance staff has not started browsing the new operation result file, the important part is extracted (step 3080), and the extracted important part is abstracted based on the abstraction rule (step 3090). The abstraction rule is input in advance from the maintenance support setting UI 2005, and describes a rule for converting the description items of the operation result file for each type of content described.
Next, the recovery work content is received from the recovery work input UI 2003 (step 3100). Up to this point, the maintenance staff finishes analyzing the cause of the failure, planning and executing the recovery work, and records the content of the recovery work in the maintenance target apparatus 1000 from the recovery work input UI 2003.

  The maintenance support program 2009 records the maintenance work content for the failure signal in the maintenance history database 1103 as a maintenance work history (step 3110).

  FIG. 8 is a data structure diagram showing an example of the maintenance work history table stored in the maintenance history database 1103 in step 3110. In the maintenance work history 8000, the work start date 8030 of the maintenance work acquired from the device ID 8010, the device type 8020, and the login information of the maintenance staff as the device information acquired from the maintenance target device 1000 or input by the maintenance staff. , The maintenance person ID 8040 indicating the information of the maintenance person who performed the maintenance work, and the maintenance person name 8050 are stored. In addition, a failure signal ID 8060, a failure signal content 8070, and a maintenance staff comment 8080 regarding the failure are stored as information on a failure that has occurred in the apparatus during the maintenance work. Further, as the information regarding the content of the maintenance work, the important part information 8200 extracted by the maintenance support program 2009 includes only the value stored in the number of important parts 8110. Further, the operation result file information 8300 output by the apparatus includes the file ID 8310 and the file path 8320 of the operation result file only for the values stored in the operation result file number 8120. Further, the recovery work information 8400 stores a recovery date and time 8410, a recovery work result 8420, and a work content 8430 performed as the recovery work.

FIG. 4 shows a processing procedure of the maintenance support program 2009 and related components at the time of maintenance work by an unskilled maintenance person. When a failure signal is reported through the failure signal monitoring program 2007, the restoration work presentation UI 2004 The procedure for performing maintenance work upon receiving the restoration work is shown.
When the maintenance staff accepts a recovery work presentation request from the recovery work presentation UI 2004, the maintenance support program 2009 acquires the current fault signal and the latest operation result file group from the fault signal monitoring program 2007 and the operation result file database 1102 ( Step 4010). In the failure signal / operation result file acquisition process, the latest information is automatically acquired from the maintenance target device 1000. However, the failure signal and the operation result file may be received from the recovery work presentation UI 2004. good.

  Next, an element having the same failure signal is searched from the maintenance work history in the maintenance history database 1103, and the past operation result file included in the current operation result file and the maintenance work history is included in the maintenance work history. The important point features are compared (step 4020). The comparison result is calculated as the similarity according to the proximity of the description contents of the important part.

  The above processing (step 4020) is performed for all maintenance work histories including the same failure signal as the currently generated failure signal (step 4020, step 4030).

  Based on the calculated similarity, the maintenance work history including the operation result file having the highest similarity is selected (step 4040), and the recovery work included in the maintenance work history is presented via the recovery work presentation UI 2004. (Step 4050).

  In the above example, the restoration work to be presented has the highest similarity. However, a plurality of restoration works are extracted together with the degree of similarity and presented in ascending order according to the degree of similarity on the restoration work presentation UI 2004. It is also good.

  Next, details of the processing procedures of FIGS. 3 and 4 will be described based on the flowcharts shown in FIGS. 6, 9, and 10. FIG. 6 shows an example of the abstraction details of the important part shown in Step 3090.

  First, the abstraction rule input by the maintenance staff is acquired from the maintenance support setting UI 2005 (step 6010). FIG. 7B is an example of an abstract rule input by maintenance personnel. Each abstract rule 7300 includes an application object 7310 indicating to which content the rule is applied, an operation 7320 indicating the type of processing performed at the time of abstraction, and an operation result to which the operation is applied. A target field 7330 indicating a field in the file, and a replacement addition item 7340 indicating information added when the operation adds new information such as replacement or addition.

  With respect to the abstract rule 7300 acquired in step 6010, an item applicable in an important place is searched (step 6020). With respect to the description content of the important part, a part that matches the application target 7310 in the abstraction rule 7300 is searched. For example, the search process will be described based on the non-abstracted important part table and the abstraction rule shown in FIGS. 7A and 7B. For a rule having “all” in the application target 7310, all important parts are described. Abstraction rule 7300 is applied. As for the abstract rule 7310 in which description other than “all” is described, a search rule indicating which type of description contents of which type of operation result file is applied is described. For example, “motor :: The description of “GET IO” indicates that the description relating to the acquisition of the IO value of the operation result file related to motor is applied.

  Although the application target 7310 of the abstraction rule 7300 described in the above description and FIG. 7B shows an example of searching for an application target using a unique type search rule, description may be based on a general-purpose character string search technique such as a regular expression. In addition, it may be described in a database search language such as an SQL sentence, or may be described in a programming language capable of expressing general-purpose processing such as Peal or C language, and a specific part of the contents described in the important part table Alternatively, any description format may be used as long as the description format can extract all.

  Next, the important part is converted based on the operation 7320, the target field 7340, and the replacement addition item 7350 described in the abstraction rule 7300 (steps 6030 to 6090). When the operation 7320 is “replacement”, the item in the important part described in the target field 7340 is searched (step 6040) and replaced with the replacement additional item 7350 (step 6050). If the operation 7320 is “delete”, the item in the important part described in the target field 7340 is searched (step 6060), and the item is deleted (step 6070). If the operation 7320 is “addition”, an item in the important part described in the target field 7340 is searched (step 6080), and a replacement addition item 7350 is added to the item (step 6090).

  For example, when the important part table described in FIG. 7A is abstracted by the abstraction rule described in FIG. 7B, the operation 7320 described in the first record of the abstraction rule 7300 is “delete”, so All the number of rows 7020 in the important part table corresponding to the “number of rows” described in 7330 is deleted. For example, since the operation 7320 described in the third record of the abstraction rule 7300 is “replacement”, the operation result 7020 in the important part table corresponding to “in the operation result” described in the target field 7330 is And “MOTOR_GET” described in the replacement addition item 7340. FIG. 7C shows the result of abstracting the important part table shown in FIG. 7A by the abstract rule shown in FIG. 7B.

Next, an example of the similarity calculation process shown in step 4020 will be described based on FIG. 9A. First, with respect to the current operation result file, a comparison target part is extracted based on an important part and an abstract rule included in the maintenance work history (step 9010).
Next, similarly to the current operation result file, the operation result file included in the maintenance work history also extracts the comparison target part based on the important part and the abstraction rule (step 9020).
The extraction process of the comparison target part from the operation result file shown in the above step 9010 and step 9020 will be described later with reference to FIG. 9B.

About the comparison object location of the present and past operation result file, both contents are compared, and the number of locations where the operation result content 7040 matches and the number of locations where they do not match are counted (step 9030).
The similarity of the operation result file is calculated based on the number of matches and the number of mismatches counted in step 9030 (step 9040). The similarity calculation method calculates a numerical value that is higher as the contents of the two operation result files are similar and that does not depend on the number of comparison target parts extracted in Step 9010 and Step 9020 or the number of important parts. Any method may be used, for example, a matching ratio obtained by dividing the number of matches by the number of extractions may be used as the similarity, or a value obtained by subtracting the number of mismatches from the number of matches may be used as the similarity. The degree of similarity may be calculated according to the set number of similarities increased per matching location and the number of similarities subtraction per mismatched location. In addition, the number of comparison target locations extracted from the current operation result file in step 9010 may be reflected in the similarity, and processing for subtracting the similarity may be added when the comparison target location is not found.

The details of the comparison target portion extraction processing shown in steps 9010 and 9020 will be described with reference to FIG. 9B.
First, an abstraction rule is used to abstract a line in the operation result file in the current or maintenance work history. The abstraction method is the same as the method shown in FIG. 6 (step 9510).
Next, it is determined whether or not there is an important part that matches the abstracted contents (step 9520). If they match, the line of the operation result file is stored (step 9530). The processing of step 9510, step 9520, and step 9530 is applied to all files and all lines of the operation result file (step 9540), and the line of the operation result file stored in step 9530 is used as a comparison target portion.

  In the above example, the important part extraction processing in step 3080 shows the method of counting the display time in the cause analysis UI 2002 with the browsing count 5140. However, the information used as the reference of the count is not only the display time but also the operation result file. Any information can be used as long as the information can be used to estimate the degree of importance and importance of the skilled maintenance staff to the specific location and to specify the location in the operation result file. Copying operation information, operation result files to the external storage device 1002, taking-out information of files such as printing from the printing device 1005, writing information from the input device 1004 such as a keyboard and touch panel, retrieval of description contents, etc. Use any or all of the survey information such as comparison with the operation result file It may be extracted an essential point.

FIG. 10 shows an example of processing when counting is performed using not only the display time but also the display state and file operation information in the count table addition processing shown in step 3040.
First, the maintenance work monitoring program 2008 obtains a line currently displayed on the display device 1003 or the like through the cause analysis UI 2002 (step 3030), and performs a moving process such as scrolling, in particular, the current display state. Get that. Only when the display state is stopped without scrolling, it is determined that the maintenance staff is browsing the corresponding line, and the browsing count 5140 is added (step 10020).
Next, it is determined whether or not a file operation has been performed up to the present time (step 10030). If printing has been performed from the printing apparatus 1005, the type of file to be printed and the printed range are acquired in units of lines (step 10040). The operation count 5150 or the browsing count 5140 corresponding to the printed line is added (step 10050). If the file operation is a file copy operation to the external storage device 1002, the type of file that was copied and the copied range are obtained in units of rows (step 10060), and an operation count corresponding to the copied row is obtained. 5150 or the browsing count 5140 is added (step 10070). When the file operation is another file for comparison, particularly when a past operation result file is opened, the display contents of the operation result file are acquired (step 10080), and the current operation result file and the past operation result are obtained. Differences are extracted from the display contents of the file (step 10090). On the basis of the extracted difference information, counting is performed for a line having a noticeable difference between the current operation result file and the past operation result file or a similar point. For example, in the example of FIG. 10, the number of matching parts is determined to some extent with respect to a preset threshold (step 10100), and when the number of matching parts is large, the viewing count 5140 corresponding to the line of the non-matching part, or The operation count 5150 is added (step 10110). When the number of matching parts is small, the browsing count 5140 or the operation count 5150 corresponding to the matched line is added (step 10120).

  In the above example, when the operation count is counted by the operation count 5140, the operation count value is converted by a conversion equation set in advance and added to the browsing count, thereby being used for extraction of an important part. The above conversion formula may be valid as long as it expresses which of the file operation and the browsing operation is important in the important part extraction process. For example, a linear conversion formula that multiplies an operation count by a preset coefficient. Any other monotonically increasing conversion formula may be used.

  A second embodiment will be described. Hereinafter, a maintenance work support system according to a second embodiment of the present invention will be described with reference to FIGS. FIG. 11 shows another embodiment of the present invention using a communication network. When a plurality of maintenance target apparatuses 1000 are maintained, maintenance work is supported from a remote server via the network.

  In FIG. 11, one or more maintenance target apparatuses 1000 are connected to the network 1103. Here, each maintenance target device 1000 may be the same type of device, or may be a different type of device. The maintenance work analysis server 11020 provides the same function as the maintenance support program 2009 described in the first embodiment on a processing device isolated from the maintenance target device 1000. The maintenance work history database 11020 stores the maintenance history maintenance work history 8000 in the same manner as the maintenance history database 1103 shown in the first embodiment. The difference from the maintenance work history database 11020 in the first embodiment is that the maintenance target device 1000 is different. It exists on the network 11030 independently, and accumulates maintenance work histories 8000 from a plurality of maintenance target apparatuses 1000.

  FIG. 12 is a software configuration diagram of the maintenance support server 11010 and the centralized controller 1100 in the maintenance target device 1000 in the maintenance support system for the plurality of maintenance target devices 1000 shown in FIG. The maintenance work history transmission program 12010 transmits the contents monitored by the maintenance work monitoring program 2008 and the operation result file group stored in the operation result file database 1102 in each maintenance target apparatus 1000 to the maintenance support server 11010. The recovery work reception program 12020 receives a recovery work presentation request from the recovery work presentation UI 2004 and acquires the recovery work from the maintenance support server 11010. The communication program 12030 performs data communication between the centralized controller 1100 and the maintenance support server 11010. The maintenance work support program 2009 receives the maintenance work history and the operation result file from the maintenance work history transmission program 12010 via the communication program 12030 and transmits the recovery work information to the recovery work reception program 12020. 2 is different from the software configuration diagram in the first embodiment shown in FIG. 2 in that the maintenance work support program 2009 is not executed in the centralized controller 1100, and the maintenance work history is exchanged with the remote maintenance work support program 2009 instead. The transmission program 12010 and the recovery work reception program operate.

  With the configuration of this embodiment, more past maintenance work can be performed by handling information from a plurality of maintenance target apparatuses 1000 than supporting maintenance work with information of only one maintenance target apparatus 1000. Support information can be created based on the history, and it is possible to extract important points with high accuracy and present recovery work.

  FIG. 13 is a flowchart illustrating an example of processing in the maintenance work history transmission program 2009 in the maintenance target apparatus 1000. The display history count is not counted, and the monitoring content of the maintenance work monitoring program 2008 is transmitted to the maintenance support server 11010.

When maintenance work is started by the cause analysis UI 2002 in the maintenance target apparatus 1000, a failure signal that is currently generated is acquired (step 13010), and an operation result file that is displayed through the cause analysis UI 2002 is displayed. A count table creation request corresponding to is sent to the maintenance work support server 11010 (step 13020). The maintenance work support server 11010 transmits a count table based on the creation request.
Next, the contents of the cause analysis work such as the display of the operation result file received from the maintenance work monitoring program 2008 are acquired, and the contents are transmitted to the maintenance work support server 11010 (step 13030).
The processing in step 13303 is executed until the analysis of the operation result file currently targeted is completed (step 13040), and the processing in steps 13020 to 13040 is performed until the cause analysis work for all the operation result files is completed. Execute (step 13050).
The transmission contents in the transmission process 13020 may be either the display contents or operation contents of the operation result file shown in the first embodiment. However, when the display content is transmitted, transmitting all the display content at regular intervals passes through the communication network 11030, which causes problems such as an increase in communication amount. Therefore, it is possible to transmit only the operation content that can identify the important part without transmitting the display content. FIG. 14 is a flowchart showing an example of the transmission process shown in step 13020.

The display state of the operation result file currently displayed on the cause analysis UI 14010 is acquired, and if the display state is stopped, the operation contents performed by the maintenance staff immediately before are acquired (steps 14010 and 14020). .
If the above operation is an input operation such as inputting characters from the keyboard, touching a specific part of the display from the touch panel, or writing to the display part from the digital pen, the line that is the target of the input operation is acquired. (Step 14030), the target line and the contents of the input operation are transmitted to the maintenance work support server 11010 (Step 14040).
When the above operation is a search operation in which a maintenance person inputs a search keyword to search for a line including the search keyword, and selects and displays a specific line from the search result. The line output as a result is acquired (step 14050), and the line is transmitted to the maintenance work support server 11010 (step 14060). Further, when one or a plurality of rows are selected in the search results, the selected rows are acquired (step 14070), and the rows are transmitted to the maintenance work support server 11010 (step 14080). .
When the maintenance work support server 11010 uses the same processing contents as the maintenance support program 2009 shown in the first embodiment to extract the important parts and present the recovery work, the maintenance target apparatus 1000 has a small amount of computer resources. In addition, it is possible to obtain the same effect as the configuration shown in the first embodiment.

  Although the present invention has been described with reference to the above-described embodiments, the first embodiment of the present invention provides a rule that describes a method for abstracting the description contents of the important portion in the maintenance work support device of the present invention. Based on the structure, it is classified into at least a replacement part, a deletion part, and an additional part, the replacement content is described for the replacement part, the abstraction rule describing the additional content is accepted for the additional part, and based on the abstraction rule, The important part information is created by abstracting the important part by replacing the replacement part with the above-mentioned replacement contents, deleting the corresponding part for the deleted part, and adding the additional contents for the additional part to create the important part information. This is a maintenance work support device that abstracts the location.

  According to another embodiment of the present invention, in the maintenance work support apparatus of the present invention, the new failure signal and the new operation result file are received, and the same type as the new failure signal in the maintenance work history. The maintenance work history including the failure signal is extracted, and the important part is extracted from the operation result file included in the maintenance work history by using the important part information included in the maintenance work history. For the extracted maintenance work history, the important part information is extracted from the new operation result file by using the important part information included in the maintenance work history, and the operation result file and the new operation result file included in the maintenance work history are extracted. The operation result file is compared for the extracted important points to calculate the similarity, and the method for calculating the similarity is the same type of failure signal as the new failure signal. The recovery work history is selected from the maintenance work history by extracting at least one recovery work history included in the maintenance work history having a high similarity and presenting it to the maintenance staff. This is a maintenance work support device to be presented.

  According to another embodiment 3 of the present invention, in the maintenance work support apparatus of the present invention, the added number of rows displayed in the central portion of the display area of the user interface is counted and displayed on the peripheral portion of the display device. This is a maintenance work support device that counts the output frequency of the operation result file by counting the added number of lines small.

  Furthermore, in Embodiment 4 of the present invention, in the maintenance work support apparatus of the present invention, when the new operation result file is extracted for the important part, the similarity is added when the corresponding part exists, and the new operation result file is added. When the important part information is extracted, the similarity is subtracted when the relevant part does not exist, and the similarity is manipulated when the important part information is created before the similarity is calculated by the comparison. It is a maintenance work support device.

  Further, in another embodiment of the present invention, in the maintenance work support device of the present invention, the restoration work is displayed in order of the degree of similarity, and at least one selection is received from the displayed restoration work, The maintenance work support apparatus presents the content of the recovery work by displaying the content of the selected recovery work.

  In another embodiment 6 of the present invention, in the maintenance work support device of the present invention, input of important part information is accepted, and the input important part information is extracted based on the output frequency count and operation frequency count. A maintenance work support apparatus that records important part information as important part information in the maintenance work history.

  Furthermore, another embodiment 7 of the present invention includes one or more maintenance target devices and a network that have a plurality of hardware and software, monitor the operation status up to the time of failure, and output as an operation result file A maintenance work support system that includes a maintenance work analysis server and a maintenance work history database that are respectively connected via a network, and presents maintenance support information to maintenance personnel when a failure occurs in any of the one or more maintenance target devices. The maintenance work analysis server acquires a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device, and a maintenance staff browses the operation result file acquired from the maintenance target device. The operation result file of the maintenance staff on the user interface is transmitted / received via the user interface with the maintenance staff terminal to be operated. The content of the browsing operation to the user is monitored and acquired, and the type of operation result file based on the acquired browsing operation content, the display frequency on the user interface for each line, the presence / absence of information input, the character string search operation Count at least one of the search adaptation frequency, the output frequency to the external output device, the output frequency of the difference information in the comparison operation with other files, and create the high-frequency browsing location information with a high count number The important location information created by identifying and abstracting the important location in the operation result file from the frequently viewed location information is created, the recovery work content of the failure performed by the maintenance staff is acquired, the maintenance work history database is A maintenance work history including at least the failure signal, the operation result file, the important part information, and the restoration work content is recorded and accumulated so as to correspond to the maintenance work. When the analysis server acquires a new failure signal and a new operation result file from the maintenance target device, the maintenance work history database has the same failure signal as the new failure signal acquired by the maintenance work analysis server. Correspondingly recorded maintenance work history is detected, based on the important part information in the detected maintenance work history, the new operation result file, the operation result file in the detected maintenance work history, The at least one restoration work content corresponding to the operation result file having a high degree of similarity is selected from the maintenance work history, and the maintenance work analysis database is selected by the maintenance work history database. This is a maintenance work support system that presents the contents of recovery work.

  Further, another embodiment 8 of the present invention has a plurality of hardware and software, and when a failure occurs in a maintenance target apparatus that monitors the operation status up to the time of the failure and outputs it as an operation result file A maintenance work support method for presenting maintenance support information to a maintenance staff at a time, acquiring a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device, and maintaining the acquired operation result file A maintenance staff terminal that is viewed and operated by an operator performs transmission / reception via the user interface, monitors and acquires the browsing operation contents of the operation result file of the maintenance staff on the user interface, and the acquired browsing operation contents Based on the type of operation result file, line by line, display frequency on the user interface, presence / absence of information input, search suitability in character string search operation At least one of frequency, output frequency to external output device, output frequency of difference information in comparison operation with other files, create high frequency browsing location information with high count number, and create the high frequency The important location information created by identifying and abstracting the important location in the operation result file from the browsing location information is created, the failure recovery work performed by the maintenance staff is acquired, at least the failure signal and the operation result When a new failure signal and a new operation result file are acquired from the maintenance target device, the file, the important part information, and the maintenance work history including the restoration work contents are recorded and accumulated in association with each other. The maintenance work history recorded and stored in response to the same fault signal as the fault signal is detected, and the new location information is based on the important part information in the detected maintenance work history. A similarity between the operation result file and the operation result file in the detected maintenance work history is calculated, and at least one restoration work content corresponding to the calculated operation result file having a high similarity is selected from the maintenance work history. This is a maintenance work support method that is selected and presented.

  According to another embodiment 9 of the present invention, in the maintenance work support method according to another embodiment 8 of the present invention, a search word for searching is input and an operation result including the search word as a search result for the search. Display the line of the file, acquire the line of the operation result file displayed in the search result, add the operation frequency count corresponding to the type of the operation result file and the line based on the acquired search result, This is a maintenance work support method for counting character string search operations based on the operation result file browsing and operation monitoring contents.

  Further, another embodiment 10 of the present invention is the maintenance work support method of another embodiment 8 of the present invention, wherein the operation result file is stored in an external storage medium via a recording device and stored in the external recording device. The operation result file is acquired for each type of operation result file and in units of lines, and the operation result file corresponding to the type of the stored operation result file and the line is added to determine whether the operation result file is externally output. This is a maintenance work support method for counting the output frequency.

  According to another embodiment 11 of the present invention, in the maintenance work support method according to another embodiment 8 of the present invention, at least one of the operation result files is selected, and the selected operation result file is selected. Display to the worker, accept selection of an operation result file different from the operation result file being displayed, compare the contents of both operation result files, display the difference, and from the contents of the difference This is a maintenance work support method that counts output operations in accordance with operation result file comparison operations by adding or subtracting operation frequency counts of rows corresponding to differences in operation result files that were initially displayed.

  Another embodiment 12 of the present invention is based on the structure of the important part as a rule describing a method of abstracting the description contents of the important part in the maintenance work support method of the other embodiment 8 of the present invention. Are classified into at least a replacement part, a deletion part, and an additional part, the replacement content is described for the replacement part, the abstraction rule describing the additional content is accepted for the additional part, and the replacement part is determined based on the abstraction rule. Replaces the content of the replacement, deletes the content of the deleted portion, adds the additional content of the added portion, abstracts the important portion as important portion information, and abstracts the important portion This is a maintenance work support method.

  Furthermore, according to another embodiment 13 of the present invention, in the maintenance work support method according to another embodiment 8 of the present invention, a current operation result file and a failure signal are received, and the current failure signal is included in the maintenance work history. The maintenance work history including the same type of failure signal is extracted, and the extracted maintenance work history is extracted from the operation result file included in the maintenance history using the important part information included in the maintenance history. And extracting the important part from the current operation result file using the important part information included in the maintenance history for the extracted maintenance work history, and the past operation result file included in the maintenance work history and the The current operation result file is compared with respect to the extracted important part, the similarity is calculated, and the similarity calculation means includes a failure signal of the same type as the current failure signal This is performed for all the maintenance work histories, and at least one of the recovery work histories included in the maintenance work history having a high similarity is extracted and presented to the user, so that the recovery work is selected and presented from the maintenance work history. This is a maintenance work support method.

  In addition, according to another embodiment 14 of the present invention, in the maintenance work support method according to another embodiment 8 of the present invention, the current display state of the operation result file is indicated as being moved when the operation of moving the display object is performed. It is obtained separately during a stop when the movement operation is not performed, and the change of the output frequency count is permitted only when the current display state is stopped, and when the current display state is moving, the This is a maintenance work support method for counting the output frequency of the operation result file by suppressing the change of the output frequency count.

  According to another embodiment 15 of the present invention, in the maintenance work support method according to another embodiment 8 of the present invention, the number of added lines displayed in the central portion of the display area of the user interface is counted, and the display device In this maintenance work support method, the output frequency of the operation result file is counted by counting the added number of lines displayed in the peripheral portion small.

  Furthermore, another embodiment 16 of the present invention relates to a maintenance work support method according to another embodiment 8 of the present invention, wherein when the current operation result file is extracted with respect to an important part, the corresponding part exists. Maintenance that manipulates the similarity in the important part before calculating the similarity by the comparison by subtracting the similarity when there is no corresponding part when the current operation result file is extracted for the important part This is a work support method.

  According to another embodiment 17 of the present invention, in the maintenance work support method according to other embodiment 8 of the present invention, the restoration work is displayed in order of the similarity, and at least one of the displayed restoration works is displayed. This is a maintenance work support method for accepting one selection and displaying the contents of the selected restoration work to present the contents of the restoration work.

  Another embodiment 18 of the present invention relates to a maintenance work support method according to another embodiment 8 of the present invention, wherein the difference information obtained by comparing a plurality of operation result files has a content more than a preset threshold value. When there are many matching lines, add the output frequency count corresponding to the lines whose contents do not match, and in the difference information, when the number of lines with matching contents is less than a preset threshold value In this maintenance work support method, the operation frequency count is changed based on the difference information by adding the output frequency count corresponding to the line whose contents match.

  Furthermore, another embodiment 19 of the present invention, in the maintenance work support method of the other embodiment 8 of the present invention, receives input of important part information, and the input important part information is output frequency count, operation frequency This is a maintenance work support method that records important part information extracted based on the count as important part information in the maintenance work history.

1 is a system configuration diagram of a maintenance target apparatus to which a maintenance work support apparatus according to a first embodiment is applied. FIG. 3 is a software configuration diagram of a control program and a maintenance support program according to the first embodiment. 3 is a flowchart illustrating an example of a maintenance history creation procedure according to the first embodiment. 5 is a flowchart illustrating an example of a restoration work presentation process procedure according to the first embodiment. FIG. 3 is a data structure diagram illustrating an example of a browsing location count table in the first embodiment. 9 is a flowchart illustrating details of an example of abstraction processing of a browsed portion in the first embodiment. The figure which shows an example of the important location table of the non-abstraction which concerns on the abstraction process of the browsing location in Example 1. FIG. FIG. 10 is a diagram illustrating an example of an abstraction rule related to the abstraction process of a browsing part in the first embodiment. FIG. 10 is a diagram illustrating an example of an abstracted important part table related to a browsing part abstraction process according to the first embodiment. 2 is a data structure diagram illustrating an example of a maintenance history table in Embodiment 1. FIG. 5 is a flowchart illustrating an example of a detailed procedure of a comparison process with an apparatus operation log according to the first embodiment. 9 is a flowchart illustrating an example of a detailed procedure of a process for extracting a portion to be compared with an apparatus operation log according to the first embodiment. 5 is a flowchart illustrating an example of count table addition processing according to the first embodiment. The system block diagram of the maintenance assistance system in Example 2. FIG. FIG. 10 is a software configuration diagram of a maintenance target device and a maintenance support server in the second embodiment. 9 is a flowchart illustrating an example of maintenance work transmission processing according to the second embodiment. 9 is a flowchart illustrating an example of maintenance work acquisition processing according to the second embodiment.

Explanation of symbols

1000 Maintenance target device 1001 Operation terminal 1002 External storage device 1003 Display device 1004 Input device 1005 Printing device 1006 Speaker 1100 Centralized controller 1101 Control information database 1102 Operation result file database 1103 Maintenance history database 1104 Communication IF
1200 Camera Controller 1202 Camera 1300 Sensor Controller 1302 Sensor 1400 Actuator Controller 1402 Actuator 2001 Device Operation UI
2002 Cause analysis UI
2003 Recovery work input UI
2004 Recovery work presentation UI
2005 Maintenance support setting UI
2006 Fault signal monitoring program 2007 Operation result monitoring program 2008 Maintenance work monitoring program 2009 Maintenance support program 2100 Control program 5000 Count table 7000 Critical location table 7020 Operation result 7300 Abstraction rules 7300 Abstraction rules 7600 Abstracted critical location table 8000 Maintenance Work history 1000 Maintenance target device 1100 Centralized controller 1102 Operation result file database 1103 Maintenance history database 2008 Maintenance work monitoring program 2009 Maintenance support program 8000 Maintenance work history 11010 Maintenance support server 11020 Maintenance work history database 11030 Network 12010 Maintenance work history transmission program

Claims (20)

Connect to a maintenance target device that has multiple hardware and software, monitors the operation status up to the time of the failure, and outputs it as an operation result file. When a failure occurs in the maintenance target device, to the maintenance staff A maintenance work support device for presenting maintenance support information,
Obtain a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device,
The maintenance staff browses and operates the acquired operation result file, and performs transmission / reception via a user interface,
Monitor and acquire the contents of the browsing operation to the operation result file of maintenance personnel on the user interface,
Based on the obtained browsing operation content, the type of the operation result file, for each line, the display frequency on the user interface, the presence / absence of information input, the search adaptation frequency in the character string search operation, the output frequency to the external output device , Count at least one of the output frequency of difference information in comparison operation with other files, create high-frequency browsing location information with a high count number,
Create important location information created by identifying and abstracting important locations in the operation result file from the created frequently viewed location information,
Obtain the work contents of the failure repair work performed by the maintenance personnel,
Recording and storing at least the failure signal, the operation result file, the important part information, and the maintenance work history including the restoration work content,
When a new failure signal and a new operation result file from the maintenance target device are acquired, a maintenance work history recorded and stored corresponding to the same failure signal as the acquired new failure signal is detected, and the detection is performed. On the basis of the important part information in the maintenance work history, the similarity between the new operation result file and the operation result file in the detected maintenance work history is calculated, and the operation result file with the high similarity is calculated. Select and present at least one restoration work content corresponding to the maintenance work history,
A maintenance work support device characterized by that. The maintenance work support device according to claim 1,
As a rule that describes how to abstract the description of the important part, classify it into at least a replacement part, a deletion part, and an additional part based on the structure of the important part. Accepts abstraction rules with additional content,
Based on the abstraction rule, the important part information is abstracted by substituting the replacement content for the replacement part, deleting the corresponding content for the deletion part, and adding the additional content for the additional part to abstract the important part. By creating
A maintenance work support device that abstracts the important part. The maintenance work support device according to claim 1,
Accept the new failure signal and new operation result file,
Extracting a maintenance work history including a failure signal of the same type as the new failure signal in the maintenance work history,
For the extracted maintenance work history, extract important parts from the operation result file included in the maintenance work history using the important part information included in the maintenance work history,
For the extracted maintenance work history, extract the important part from the new operation result file using the important part information included in the maintenance work history,
Compare the operation result file and the new operation result file included in the maintenance work history with respect to the extracted important part, and calculate the similarity,
The method of calculating the similarity is performed on all maintenance work histories including the same type of failure signal as the new failure signal, and at least one recovery work history included in the maintenance work history having the high similarity is extracted. And present it to maintenance personnel,
A maintenance work support apparatus, wherein a restoration work is selected and presented from the maintenance work history. The maintenance work support device according to claim 1,
Count many additions of rows displayed in the center of the display area of the user interface,
By counting the added number of lines displayed on the periphery of the display device small,
A maintenance work support apparatus that counts the output frequency of the operation result file. The maintenance work support device according to claim 1,
When the new operation result file is extracted for the important part, if the corresponding part exists, add the similarity,
By subtracting the similarity when there is no relevant part when extracting the new operation result file for the important part,
A maintenance work support apparatus, wherein the similarity is operated when creating important part information before calculating the similarity by the comparison. The maintenance work support device according to claim 1,
Display recovery work in order of the degree of similarity,
Accept at least one selection from the displayed recovery operations;
By displaying the contents of the selected recovery work,
A maintenance work support device that presents the contents of the restoration work. The maintenance work support device according to claim 1,
Accept important part information input,
Record the input important part information as the important part information in the maintenance work history together with the important part information extracted based on the output frequency count and the operation frequency count.
A maintenance work support device characterized by that. Maintenance work analysis server that has multiple hardware and software, and that connects to one or more maintenance target devices that monitor the operation status up to the time of the failure and output it as an operation result file, respectively, via the network and maintenance A maintenance work support system comprising a work history database and presenting maintenance support information to maintenance personnel when a failure occurs in any of the one or more maintenance target devices;
The maintenance work analysis server acquires a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device, and maintenance is performed by a maintenance person browsing and operating the operation result file acquired from the maintenance target device. A user terminal and a user interface are used for transmission and reception, and monitoring operation contents of the maintenance staff on the user interface on the operation result file are monitored and acquired. Based on the acquired browsing operation contents, the operation result file For each type and line, display frequency on the user interface, presence / absence of information input, search adaptation frequency in character string search operation, output frequency to external output device, output frequency of difference information in comparison operation with other files Count at least one of, create high-frequency browsing location information with a high count number, Create an important place information that was created to identify and abstract the important part of the operating result in the file, to get the restoration work of failure that the maintenance personnel went,
The maintenance work history database records and accumulates maintenance work history including at least the failure signal, the operation result file, the important part information, and the restoration work content,
When the maintenance work analysis server acquires a new failure signal and a new operation result file from the maintenance target device, the maintenance work history database is the same as the new failure signal acquired by the maintenance work analysis server. The maintenance work history recorded and stored in response to the failure signal is detected, and the new operation result file and the operation in the detected maintenance work history are detected based on the important part information in the detected maintenance work history. Calculate the similarity with the result file, select at least one restoration work content corresponding to the operation result file with the calculated high similarity from the maintenance work history,
The maintenance work analysis server presents the recovery work content selected by the maintenance work history database;
Maintenance work support system characterized by that. Maintenance work that presents maintenance support information to maintenance personnel when a failure occurs in a maintenance target device that has multiple hardware and software, monitors the operation status up to the time of the failure, and outputs it as an operation result file A support method,
Obtain a failure signal and an operation result file indicating information on a failure occurring in the maintenance target device,
The maintenance staff browses and operates the acquired operation result file, and performs transmission / reception via a user interface,
Monitor and acquire the contents of the browsing operation to the operation result file of maintenance personnel on the user interface,
Based on the obtained browsing operation content, the type of the operation result file, for each line, the display frequency on the user interface, the presence / absence of information input, the search adaptation frequency in the character string search operation, the output frequency to the external output device , Count at least one of the output frequency of difference information in comparison operation with other files, create high-frequency browsing location information with a high count number,
Create important location information created by identifying and abstracting important locations in the operation result file from the created frequently viewed location information,
Obtain the work contents of the failure repair work performed by the maintenance personnel,
Recording and storing at least the failure signal, the operation result file, the important part information, and the maintenance work history including the restoration work content,
When a new failure signal and a new operation result file from the maintenance target device are acquired, a maintenance work history recorded and stored corresponding to the same failure signal as the acquired new failure signal is detected, and the detection is performed. On the basis of the important part information in the maintenance work history, the similarity between the new operation result file and the operation result file in the detected maintenance work history is calculated, and the operation result file with the high similarity is calculated. Select and present at least one restoration work content corresponding to the maintenance work history,
A maintenance work support method characterized by the above. The maintenance work support method according to claim 9,
Enter a search term to search,
As a search result for the search, a line of an operation result file including a search word is displayed.
Get the line of the operation result file displayed in the search result,
Based on the acquired search results, the type of the operation result file, the operation frequency count corresponding to each line is added,
A maintenance work support method, comprising: counting a character string search operation based on the operation result file browsing and operation monitoring contents. The maintenance work support method according to claim 9,
Storing the operation result file in an external storage medium via a recording device;
The content stored in the external recording device is acquired by the type of the operation result file, in units of lines,
By adding the operation frequency count corresponding to the type and line of the saved operation result file,
A maintenance work support method, wherein the output frequency is counted according to the presence or absence of external output of the operation result file. The maintenance work support method according to claim 9,
Selecting at least one of the operation result files, and displaying the selected operation result file to the worker;
Accepting selection of an operation result file different from the operation result file being displayed, comparing the contents of both operation result files, and displaying the difference,
By adding or subtracting the operation frequency count of the line corresponding to the difference of the operation result file that was initially displayed from the content of the difference,
A maintenance work support method, characterized in that an output operation is counted according to an operation result file comparison operation. The maintenance work support method according to claim 9,
As a rule that describes how to abstract the description of the important part, classify it into at least a replacement part, a deletion part, and an additional part based on the structure of the important part. Accepts abstraction rules with additional content,
Based on the abstraction rule, the replacement part is replaced with the replacement content, the deletion part is deleted with the corresponding content, the additional part is added with the additional content to abstract the important part and the important part information By doing
A maintenance work support method characterized by abstracting the important part. The maintenance work support method according to claim 9,
Accept the current operation result file and failure signal,
Extracting a maintenance work history including a failure signal of the same type as the current failure signal in the maintenance work history,
For the extracted maintenance work history, extract the important part from the operation result file included in the maintenance history using the important part information included in the maintenance history,
For the extracted maintenance work history, extract the important part from the current operation result file using the important part information included in the maintenance history,
Compare the past operation result file included in the maintenance work history and the current operation result file for the extracted important part, and calculate the similarity,
The similarity calculation means is performed for all maintenance work histories including a failure signal of the same type as the current failure signal, and at least one recovery work history included in the maintenance work history having a high similarity is extracted. By presenting it to the user,
A maintenance work support method comprising selecting and presenting a restoration work from the maintenance work history. The maintenance work support method according to claim 9,
The current display state of the operation result file is obtained by distinguishing between the movement operation of the display object being performed and the stop operation in which the movement operation is not performed,
By permitting the change of the output frequency count only when the current display state is stopped, and suppressing the change of the output frequency count when the current display state is moving,
A maintenance work support method, wherein the output frequency of the operation result file is counted. The maintenance work support method according to claim 9,
Count many additions of rows displayed in the center of the display area of the user interface,
By counting the added number of lines displayed on the periphery of the display device small,
A maintenance work support method, wherein the output frequency of the operation result file is counted. The maintenance work support method according to claim 9,
When the current operation result file is extracted for the important part, if the corresponding part exists, add the similarity,
When the current operation result file is extracted for the important part, if the corresponding part does not exist, the similarity is subtracted,
A maintenance work support method, wherein the similarity in an important part is manipulated before the similarity is calculated by the comparison. The maintenance work support method according to claim 9,
Display recovery work in order of the degree of similarity,
Accept at least one selection from the displayed recovery operations;
By displaying the contents of the selected recovery work,
A maintenance work support method characterized by presenting recovery work contents. The maintenance work support method according to claim 9,
In the difference information comparing a plurality of operation result files, if there are more lines whose contents match than the preset threshold, add the output frequency count corresponding to the lines where the contents do not match,
In the difference information, when the number of matching lines is less than a preset threshold, by adding an output frequency count corresponding to the matching lines,
A maintenance work support method, wherein an operation frequency count is changed based on the difference information. The maintenance work support method according to claim 9,
Accept important part information input,
Record the input important part information as the important part information in the maintenance work history together with the important part information extracted based on the output frequency count and the operation frequency count.
A maintenance work support method characterized by the above.

Images