JP2007226532A - Maintenance information sharing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To organize a system capable of cross-sectionally sharing defect information and maintenance information in different fields, which are individually generated by each piece of mechanical equipment, by allowing an equipment user and a vendor user to share the maintenance information via the Internet. <P>SOLUTION: A defect database 101 composed of equipment data and predicate data, a cause database 102 composed of object data and predicate data, and a maintenance database 103 composed of element data and predicate data, are associated with each other in an ontology data structure. The maintenance information sharing system includes a means for collectively transmitting the equipment data and predicate data the object data, and predicate data and the element data, and predicate data, which are extracted from each database, with the use of an HTML language. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention is a maintenance information sharing system that can display failure information of equipment and maintenance information for the failure on a shared platform, and in particular, equipment users and vendor users share failure events via the Internet. It is a system that can.

  In the manufacturing industry where a wide variety of equipment machines operate, many equipment machines introduced during the high growth period have deteriorated over time. Therefore, the demand for equipment maintenance technology is increasing. Such maintenance technology plays an important role in eliminating equipment failures and quality defects and improving productivity. For example, when a facility failure occurs, productivity can be improved by repairing a facility that has failed at the failure site quickly and properly.

  As a result of diversification of corporate management, there are many cases where the equipment industry type and the processing / assembly type coexist in one factory. Further, the manufacturing process is composed of a combination of unit operations such as pulverization, dissolution, reaction filtration, adsorption, concentration, crystallization, separation, molding, drying and cooling, and conveyor transportation of various articles. Among them, equipment machinery includes stationary equipment such as towers, tanks, heat exchangers, kettles, and heating furnaces, rotary machines such as pumps, compressors, motors, and turbines, pipelines that connect them, and control system electricity. It is configured by instrumentation.

  At the manufacturing site where these various equipment machines are handled, a field engineer is assigned to each device to maintain vertically divided equipment. For example, when an oil leak occurs in a workpiece in robot welding, it is necessary for a field engineer of the robot welding machine to identify a defective part and perform appropriate repairs. In reality, these tasks are actually left to the experience and skill of field engineers. Quality maintenance management is often introduced in order to smoothly perform such equipment maintenance. Today, with regard to management systems related to equipment maintenance, many business establishments in large corporations are developing quality maintenance management while maintaining equipment and accumulating maintenance know-how using maintenance sheets. (See Non-Patent Document 1).

  Such quality maintenance management accumulates materials handwritten or typed on maintenance sheets, etc., and when similar problems occur, search for past materials using human resources and perform maintenance according to the materials. Currently, there is a difference in the contents of materials for each company and department regarding the efforts for the accumulation and effective use of materials.

  In addition, in such maintenance management that makes full use of human resources, it is important to secure skilled field engineers, but in recent years there has been a problem of aging and decreasing the number of veteran employees. Also, maintenance management is carried out in a vertical division at each site, and standardization of materials such as maintenance sheets is not sufficient, and it is difficult to share the accumulated maintenance information at different sites within the same company. . In particular, there is no standardization of materials between different companies.

In view of the difficulty of searching for a coping method according to a failure location and situation by making full use of such human resources, a failure diagnosis device based on a target ontology database has been created (see Patent Document 1).
However, the problem of sharing maintenance information between equipment users and vendor users and cross-sectional sharing in different fields have not been solved.
New TPM deployment program for industrial innovation (Established by Japan Plant Maintenance Association) No. 2000-322125

An object of the present invention is to display defect information and maintenance information on a shared platform in order to perform quick and qualified facility maintenance.
It is another object of the present invention to construct a system that allows facility users and vendor users to share maintenance information via the Internet.
Another object of the present invention is to construct a system that can share defect information and maintenance information that have been created for each machine equipment field across different fields.

The present invention employs the following configuration in order to solve the above problems.
(1) In the present invention, a failure database composed of equipment data and predicate data, a cause database composed of target data and predicate data, and a maintenance database composed of element data and predicate data are associated in an ontology data structure, A maintenance information sharing system comprising means for collectively transmitting facility data and predicate data, target data and predicate data, element data and predicate data extracted from each database in an HTML language.

  Here, the defect database manages arbitrary equipment data and predicate data. The equipment data is an item indicating a machine facility or a part of the machine facility. For example, it is a proper noun such as “cutting machine”, “demolding machine”, “press machine”, “pump”, “fan” and the like. The predicate data is an item indicating a predicate related to a defect in the facility data. For example, predicates such as “burn out”, “corrode”, “leak liquid”, “damage”, “vibrate”, and the like. These items are composed of items arbitrarily selected from materials (for example, maintenance sheets, etc.) relating to maintenance created in the past, and each is assigned an individual ID number to form the first layer.

  The cause database manages arbitrary target data and predicate data. The target data is an item indicating an event that is a main cause of the failure of the mechanical equipment. For example, “foreign matter”, “overload”, “overcurrent” and the like can be mentioned. The predicate data is an item indicating a predicate related to the cause of failure. For example, “entrain”, “occur”, “decrease”, and the like. These items are composed of items arbitrarily selected from maintenance sheets or the like prepared in the past, and each is assigned an individual ID number to constitute the second layer. These subdivide fault causes related to faults managed in the fault database, target data is composed of items indicating the subdivided subjects, and predicate data is composed of subdivided predicates.

  Further, the maintenance database manages arbitrary element data and predicate data. The element data is an item indicating an event that is a main body of maintenance measures against a failure. For example, “thermal relay”, “gear ratio”, “hydraulic pressure” and the like. The predicate data is an item indicating a predicate related to maintenance measures. For example, “change setting”, “change”, “set to appropriate value”, and the like. These items are composed of items arbitrarily selected from maintenance sheets or the like prepared in the past, and each is assigned an individual ID number to constitute the third layer. These items are maintenance data corresponding to each combination of the target data and predicate data in the cause database, and are generated, for example, by a combination of “overload” (target data) + “occurs” (predicate data) in the cause database. It consists of nodes that serve as maintenance measures against the cause of failure. In the case of this combination, of the solution means “change the gear ratio” as the maintenance measure, “gear ratio” is stored as element data and “change” is stored as predicate data in the maintenance database. Keep it.

The above defect database (first level) and cause database (second level)
The data stored with the ID number assigned to the defect database (first level) and the data stored in the cause database (second level) related to the data are associated with the ID number in a tree shape. It has been. Similarly, the data stored in the cause database (second hierarchy) and the data stored in the maintenance database (third hierarchy) are associated with an ID number in a tree shape. Through such association, each database constitutes an ontology data structure.

In this way, defect information is divided into a high-level concept (defect database) and a middle-level concept (cause database), and each is classified into a subject and a predicate, and further maintenance information is classified into two types: a subject and a predicate. Materials such as maintenance sheets that were created separately can be standardized.
(2) The present invention is a means for extracting one piece of equipment data and predicate data from the defect database based on the defect information, and a means for extracting one or more target data and predicate data associated with the data from the cause database. Means for extracting one element data and predicate data associated with the data extracted from the defect database and the cause database from the maintenance database, and means for collectively transmitting all the data extracted from each database in the HTML language The maintenance information sharing system described in (1) above.
(3) The maintenance information according to (1) or (2), wherein the present invention includes means for extracting one element data and predicate data associated with the data extracted from the cause database from the maintenance database. It is a shared system.

  As described above, the defect database (first hierarchy) and the cause database (second hierarchy) are associated in a tree shape, and the cause database (second hierarchy) and the maintenance database (third hierarchy) are associated. . Accordingly, the defect database and the maintenance database are not necessarily associated with each other, and the data that is not associated with the maintenance data in which the defect database (first hierarchy) and the maintenance database (third hierarchy) were created in the past is extracted. , There is a possibility of being sent in batches in HTML language.

With such a function, it is possible to share defect data created for each field of machinery and equipment and maintenance data as countermeasures across the data.
(4) The present invention is characterized in that the equipment user client terminal, the vendor client terminal, and the server are connected via the Internet, and the extracted failure event is displayed on the monitor of the client terminal. It is a maintenance information sharing system described in any one of 1) to (3).
(5) The present invention deletes the target data and \ or predicate data in the cause database, or the element data and \ or predicate data in the maintenance database by a signal input from the equipment user client terminal and \ or the vendor client terminal. The maintenance information sharing system according to claim 4, further comprising means that can be changed and added.
With such a system, a plurality of equipment users and vendor users as well as a plurality of equipment users and vendor users can share maintenance information via the Internet.
(6) Displaying a radar chart composed of effects, ease, reliability, immediate effect, cost, introduction, introduction rank, seriousness related to maintenance data stored in the database on the same monitor. The maintenance information sharing system described in (4) above

The inventions according to claims 1 to 3 have an effect that the defect information and the maintenance information can be displayed on the shared platform in the HTML language. In addition, there is an effect that defect information and maintenance information created for each machine equipment field can be shared across different fields.
Further, the invention according to claims 4 to 6 has an effect of constructing a system in which facility users and vendor users can share trouble events via the Internet.

An embodiment of a maintenance information sharing system according to the present invention will be described. The following data structure and the like are examples, and other data structures and programs may be used. That is, the following embodiments do not limit the scope of the invention described in the claims.
FIG. 1 is a block diagram showing a functional configuration in the server 1 of the maintenance information sharing system. In FIG. 1, 10 indicates a database unit, and 20 indicates a data processing unit.
A signal related to defect information input from an operation unit such as a keyboard (not shown) is input into the server 1 via the WWW server 36. The defect information to be input is composed of items indicating a machine facility in which a defect has occurred or a part of the machine facility and a predicate item indicating a defect state. The input signal is input to the data processing unit 20 via the interface unit 30. In the data processing unit 20, first, facility data and predicate data are specified by the parameter analysis unit 201. Next, the equipment data and the predicate data specified by the parameter search unit 202 are searched from the defect database (S1). As a result of the search, facility data and predicate data that are compatible with the defect data extraction unit 203 are extracted from the defect database (S2). The extracted defect data is transmitted to the display data creation unit 34 (S3). The extracted defect data is transmitted to the defect data search unit 204, and the target data and predicate data related to the defect data are searched from the cause database 102 (S4). As a result of the search, target data and predicate data matching the cause data extraction unit 205 are extracted (S5). The extracted cause data is transmitted to the display data creation unit 34 (S6). The extracted cause data is transmitted to the cause data search unit 206, and element data and predicate data related to the cause data are searched from the maintenance database 103 (S7). As a result of the search, element data and predicate data suitable for the maintenance data extraction unit 207 are extracted (S8). The extracted maintenance data is transmitted to the display data creation unit 34 (S9).

  The display data creation unit 34 transmits the facility data and the predicate data, the target data and the predicate data, the element data and the predicate data extracted from each of the databases 101, 102, and 103 to the WWW server 36 in a batch using the HTML language. To do.

Next, the data structure in each database in the database 10 will be described with reference to FIG.
As shown in FIG. 2, nodes 111, 121, 131,... Composed of two types of data, equipment data and predicate data, are stored in the defect database 101. Each node is assigned an individual ID number. For example, “AABB00000000”. The “AA” part is assigned an individual ID number for the equipment data. For example, “01”, “02”... The part “BB” is assigned an individual ID number for the predicate data. The defect database 101 constitutes a first hierarchy of a tree-like data structure by accumulating ID numbers assigned by the above method.

Next, in the cause database 102, nodes 112, 122, 132, 142, 152, 162,... Composed of two types of data, target data and predicate data, are stored. Each node has an individual ID number “0000 CCDD0000”. The “CC” part is assigned an ID number for the target data. Here, an ID number for the predicate data is assigned to the “DD” part. These ID numbers are associated with defect data. For example, if the ID number of the failure data node 111 is “0101 CCD DEEFF”, the ID number of the node 112 in the cause database 102 is “01010101EEFF”, for example. And the cause database 102 comprises the 2nd hierarchy of a tree-like data structure by accumulating ID number attached | subjected by said method. In FIG. 2, each node in the cause database 102 is branched into three from each node in the defect database 101. However, depending on the number of associated data, the number is one or four or more. Sometimes.

  Next, in the maintenance database 103, nodes 113, 123, 133, 143, 153, 163, 173, 183, 193, 104, 114, 124,... Composed of two types of element data and predicate data are stored. It is remembered. Each node is assigned an individual ID number “00000000EEFF”. Here, the “EE” part is given an ID number for the element data. The “FF” part is given an ID number for the predicate data. These ID numbers are associated with cause data. For example, if the ID number of the node 112 of the cause data is “01010101EEFF”, the node 113 in the maintenance database 103 is “AABB01010101”, for example. The maintenance database 103 constitutes a third hierarchy of a tree-like data structure by accumulating the ID numbers assigned by the above method. In FIG. 2, the maintenance data is branched into two from the cause data. However, depending on the number of associated data, there may be one or no data.

Furthermore, the ID number in the cause database 102 (second layer) may be duplicated. In particular, if the items of the target data and the predicate data are the same, it is preferable to attach a common ID number. There is no common ID number in one tree, but there may be a common ID number with other trees (those having different first layers). In such a case, the third hierarchies of different trees can be associated with each other using an ID number common to the second hierarchies. For example, by assigning the ID number of the node 132 as “AABB0103EEFF” and the ID number of the node 142 as “AABB0103EEFF”, the second hierarchy can be associated.
As described above, an ontology data structure is configured by associating common attribute values. Note that the above ID numbering method is an example, and may be three digits, or may be related by other methods, and does not limit the present invention.

FIG. 3 is a conceptual diagram for explaining one embodiment of the invention according to claim 4.
The maintenance information sharing system according to the present embodiment includes a server 1 having the data structure described above, equipment user clients 41, 41 ′, 41 ″... Having a WWW browser, and vendor user clients 42, 42 ′. , 42 ″, etc. are formed by being connected via the Internet. By using the Internet, the facility user and the vendor user can share maintenance information using the WWW server 36 as a platform. In particular, since the data structure in the database in the server 1 is associated in the second hierarchy as described above, the defect information and the maintenance information that have been created individually for each machine facility field are cross-sectional in different fields. Can be shared.

Next, information displayed on the WWW browser will be specifically described.
FIG. 4A is a simplified material of data created by quality maintenance management. FIG. 4B is an example of a data sheet stored in the database. Here, a case where the motor is burned out will be described as an example. Conventionally, many maintenance sheets have been prepared for motor burnout. However, at present, these maintenance sheets are created in a language / format unique to each site. Therefore, the trouble that actually becomes a problem is divided into facility data and predicate data extracted from the maintenance sheet and stored in the trouble database. Here, “motor” is stored as equipment data, “burnout” is stored as predicate data, and individual ID numbers are stored.

Next, the cause of the failure is extracted from the description on the maintenance sheet. When the cause is explained in a hierarchy, it is divided into target data and predicate data for each hierarchy and stored in the cause database. From the maintenance sheet shown in FIG. 4 (a), the cause of the failure is divided into hierarchies i) due to rubber entrainment, ii) “overload occurs”, and iii) “overcurrent flows”. Next, i) is changed to FIG.
As shown in b), “foreign matter” is stored as target data, and an ID number is assigned and stored as “entry” as predicate data. Similarly, ii) is divided into the target data “overload” and the predicate data “occurs”, and iii) is divided into the target data “overcurrent” and the predicate data “occurs” and stored with an ID. The ID numbers of the first hierarchy of these data are all common and form a tree-like data structure branched into three.

  Next, a procedure for assigning ID numbers to maintenance information in order to construct a maintenance database will be described. The maintenance information in the maintenance sheet of FIG. 4A is “change the setting of the thermal relay”. Since the thermal relay suppresses overcurrent, the second hierarchy is the same as the ID number of iii) described above, and an individual ID number is assigned to the third hierarchy and stored in the maintenance database.

  By analyzing and storing the maintenance sheets collected in the past in this way, the defect database, the cause database, and the maintenance database are constructed in a tree form. What is important here is that data related to a wide variety of machine equipment (for example, cutting machines, demolding machines, press machines, pumps, fans, conveyors, etc.) and company maintenance sheets are stored separately in equipment and predicate data. It is. These data are stored in the database in a tree form, but are systematically related in the second hierarchy and have an ontology data structure.

FIG. 5 is an initial screen diagram showing an image displayed on the monitor of the facility user client and the vendor user client.
First, defect information is input according to the initial screen as shown in FIG. At this time, the target data column (blank “subject (what)” on the left side of the screen) and the predicate data column (blank “predicate (what / what)” on the right side of the screen) and “ga” and “ ”Will be entered respectively.

  After inputting such defect information, the defect database is accessed via the interface unit 30 in FIG. 1 by clicking the search icon displayed at the lower center of the screen, and the matching facility data and predicate data are extracted. . Thereafter, in the steps described above, target data and predicate data suitable for the cause data extraction unit 205 are extracted, and element data and predicate data suitable for the maintenance data extraction unit 207 are extracted. The display data creation unit 34 collectively transmits the facility data and predicate data, the target data and predicate data, the element data, and the predicate data extracted from each of the databases 101, 102, and 103 to the WWW server 36 in the HTML language.

FIG. 6 is a diagram showing an image image in which all data transmitted to the WWW server 36 are displayed on the browsers of the facility user client and the vendor user client.
As shown in FIG. 6, the facility data and predicate data extracted from the defect database as the first hierarchy are displayed in the lower left column, and the target data and predicate data extracted from the cause database are displayed in the upper left column. The extracted element data and predicate data are displayed in the upper right column.

Here, maintenance data corresponding to the three types of data displayed in the upper right column is displayed corresponding to the upper right, but there is a blank space in the upper right column. This indicates that maintenance data related to the cause data is not stored. This means that there is currently no means for maintaining the cause of the failure, but if there is a means for the vendor user who is browsing the browser to maintain using his / her own technology, he / she can write at any time. It is also expected that vendor users who have viewed the blanks will develop new maintenance methods in-house.
Furthermore, as shown in FIG. 6, on the right side, a radar chart composed of effects, ease, reliability, immediate effect, cost, introduction, introduction rank, and seriousness related to maintenance data stored in the database in advance is displayed. You may display. Here, the radar chart is created by a known technique.

  The data displayed in FIG. 6 is cause data and maintenance data related to the malfunction “motor burns out”. Here, when the cause data common to the second hierarchy of the tree relating to the mechanical equipment other than the motor is included, the cause data can be displayed and related maintenance data can be displayed.

  For example, there is a place “overload occurs” in the upper left column of FIG. 6, but the corresponding maintenance data column is blank in this place. When the “overload occurs” column is clicked and data common to the second hierarchy exists in the tree of other machine equipment, a screen as shown in FIG. 7 is displayed on the monitor. FIG. 7 is data relating to a tree relating to “conveyor breakage” as, for example, defect data. When the second level of the tree and the second level of the data related to the tree related to the “motor burnout” are common, the maintenance data regarding the breakage of the conveyor is displayed as shown in the screen of FIG. Since such data is different in mechanical equipment, conventionally, maintenance data could not be shared across the board. However, if the maintenance information sharing system according to the present embodiment is used, maintenance data for conveyor breakage can be referred to for the problem of motor burnout.

It is a block diagram which shows an example of a function structure of a maintenance information sharing system. It is a conceptual diagram which shows the data structure in a database. It is a conceptual diagram for demonstrating one embodiment of the invention which concerns on Claim 4. (A) is a simplified data example of data created by quality maintenance management. (B) is an example of the data sheet memorize | stored in a database. It is a monitor figure which shows the image displayed on the monitor of an equipment user client and a vendor user client. Displayed on the browsers of equipment user clients and vendor user clients It is a monitor figure which shows the screen image on which the data with which a 2nd hierarchy is common were displayed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Server 10 Database part 20 Data processing part 30 Interface part 34 Display data creation part 36 WWW server 101 Defect database 102 Cause database 103 Maintenance database

Claims (6)

The defect database consisting of equipment data and predicate data, the cause database consisting of target data and predicate data, and the maintenance database consisting of element data and predicate data are associated with the ontology data structure, and the equipment extracted from each database A maintenance information sharing system comprising means for collectively transmitting data and predicate data, target data and predicate data, element data and predicate data in an HTML language. Based on the defect information, means for extracting one piece of equipment data and predicate data from the defect database, means for extracting one or more target data and predicate data associated with the data from the cause database, and from the defect database and the cause database And a means for extracting one element data and predicate data associated with the extracted data from a maintenance database, and a means for transmitting all data extracted from each database in a batch in the HTML language. The maintenance information sharing system according to claim 1. 3. The maintenance information sharing system according to claim 1, further comprising means for extracting one element data and predicate data associated with the data extracted from the cause database from a maintenance database. The facility user client terminal, the vendor client terminal, and the server are connected via the Internet, and the extracted malfunction event is displayed on the monitor of the client terminal. Maintenance information sharing system described in. The target data and \ or predicate data in the cause database, or the element data and \ or predicate data in the maintenance database can be deleted / changed / added by a signal input from the equipment user client terminal and \ or the vendor client terminal. The maintenance information sharing system according to claim 4, further comprising means capable of performing the maintenance. A radar chart composed of effects, ease, reliability, immediate effect, cost, introduction, introduction rank, and seriousness related to maintenance data stored in advance in a database is displayed on the same monitor. The maintenance information sharing system according to claim 4.

Images