JP2777278B2 - Information retrieval method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention searches information required by a user from an information bank such as a database in an information processing field using a computer and outputs the information. The present invention relates to an information search method and an apparatus therefor.

(Prior Art) A method of searching for information from an information bank such as a database management system configured using a computer is based on data items (for example, name, age, etc.) registered in a database and actual data (for example, , Shimizu, 33 years old, etc.)
With an AND search (for example, “name = Shimizu” and “age = 3
0, etc.), OR search (for example, “Name = Shimizu” or “Age = 30”, etc.), or condition-designated search (for example, “Age> 30”, etc.), or fuzzy designation search (for example, The data required by the user is searched from the database by a search condition setting method combining “name = clean *”; *, etc.).

In addition, the method of outputting the search results is to set the data items to be output in advance by setting the above search conditions, and to set the data that matches the search conditions or the contents of the data items related to the data, that is, the data. It can be output in table format.

(Problems to be Solved by the Invention) However, 1) The user's thinking is often ambiguous, and appropriate search conditions cannot be set.

That is, there are various combinations of the setting of the search condition, and it is often difficult for the user to set the search condition that matches his / her thought.

2) The conventional method can set only partial conditions of the searcher's thinking and search intention.

3) In the conventional method, it is not possible to identify and classify the importance and priority of information necessary for a searcher's decision from search information.

For this reason, only the unique information that matches the search condition is provided.

 There were problems and issues to be improved.

Originally, retrieval of information from a database involves various requests, such as 1) extracting information that meets conditions, and 2) identifying and extracting information that assists a searcher in making a decision. In the method (1), only the level 1) is used, and the effective use of the information bank is limited.

This will be described with reference to the examples shown in FIG. 11 and FIG.

FIG. 11 is an example in which the outline of the content of data management by a relational database is schematically illustrated. In the conventional search example, for example, when searching for the name, age, and affiliation of a person who smokes and who is 35 years or older, the search condition is, for example, “smoking status = Yes” AND “age> 35” / “Output = name,
Age, affiliation ", etc. (This expression is shown as an example of a general format because the expression differs depending on the computer and database system.) And, as a result, FIG. 12 The output information of the search result as shown in FIG.

By the way, this user often searches for a certain purpose, for example, in this example, selection of a subject for a lung examination, and the selection criterion is “a person who needs a screening”. If there are a plurality of persons, such as "persons who can ask questions" and "persons who do not need medical examinations", setting of search conditions becomes complicated, and information that satisfies the user's intention is extracted in one search. You can't do that. This becomes more difficult as the conditions for judgment and selection increase.

The present invention has been made in view of such a point, and can systematically and appropriately express a user's thinking and a search purpose, and can simultaneously support a searcher's thinking and its decision making. It is an object of the present invention to provide a search method and an apparatus therefor.

[Configuration of the invention]

(Means for Solving the Problems) In order to achieve the above-mentioned object, the present invention develops a request such as a searcher's information search purpose and a decision-making procedure stepwise by branching based on individual judgment conditions. Logic connected
A step of expressing and setting as search conditions in a tree format;
A method of retrieving information from a database management device, comprising: organizing target information retrieved at each final branch point or in the middle of a logic tree according to the purpose of discriminating and identifying each branch point, and outputting the information. Is what you do.

Further, the present invention provides a database management device in which search information is collected and accumulated, an information calculation process / storage device connected to the database management device, which processes and stores search information, and an information calculation process / storage device. An information input device connected to the device for inputting information; and an information input / output display device connected to the information calculation processing / storage device for displaying input / output information of the information. , A thoughtful search function that expresses search conditions in the form of a logic tree, organizes target information searched for at each final branch point or in the middle of a branch according to the purpose of discriminating and identifying each branch point, and outputs information. Another object of the present invention is to provide an information retrieval device characterized by having the information retrieval device.

(Operation) According to the present invention, the setting of a search condition in the form of a logic tree or the like according to various thoughts of a user is performed by an information input / output display device, for example, a CRT screen, via an information processing / storage device of a computer. Using information input devices such as a mouse and a keyboard, processing of setting and correcting search conditions and information to be searched and output is performed through information input / output processing of graphics and characters, and also through information arithmetic processing / storage devices. Then, information search from the database management device is performed. The search condition and search output target information created by the thoughtful search function are registered in the information operation processing / storage device or called from the information operation processing / storage device by applying the thoughtful search function. Furthermore, the information searched by the thoughtful search function is
It is arranged in the form of a chart or the like and displayed on the information input / output display device via the information processing / storage device.

(Example) Hereinafter, an example of the present invention is described with reference to drawings.

FIG. 1 is a block diagram showing the configuration of an information search device according to the present invention. The information search device includes an information processing / storage device 2 built in the computer 1 and a database management device 3 connected to the device 2 and collecting and accumulating search information.

The information processing / storage device 2 expresses a searcher's request for information search purpose and decision making procedure as a search condition in the form of a logic tree in which branches based on individual judgment conditions are tomographically expanded and connected. In addition to the setting, a thoughtful search function is provided for organizing target information searched at each final branch point or in the middle of a branch of the logic tree according to the purpose of determining and identifying each branch point and outputting the information.

The information processing / storage device 2 includes a cathode ray tube (CR)
An information input / output display device 4 for displaying an information input / output state as shown in T) and an information input device 5 such as a mouse and a keyboard are connected. An information output device 6 for printing output information, such as a printer, is connected to the information processing / storage device 2.

Using the information retrieval device having such a configuration, the eleventh
A method for searching for necessary information from a database as shown in the figure will be described.

First, on the information input / output display device 4, for example, a CRT screen,
Using the information input device 5 such as a mouse or a keyboard, information input / output processing of graphics and characters is performed. By this processing, as shown in FIG. 2, a request such as a user's information search purpose or a decision making procedure expands branches based on individual judgment conditions in a tomographic manner and searches in a form of a linked logic tree or the like. Expressed as conditions and set.

Subsequently, a search is performed using the search conditions, and the target information searched at each final branch point or in the middle of the branch of the logic tree is combined with search output target information set in advance according to the purpose of determination and identification of each branch point. Organize in table format and output information. FIG. 3 is a display example of output information searched in the middle of a branch for each purpose of judgment identification, and FIG. 4 is a display example of a search result at each final branch point.

As described above, according to the present embodiment, the output information as shown in FIG. 4 can be extracted by one search, so that the user's purpose can be satisfied and the user's decision support can be performed. .

Next, the present invention extracts the knowledge and technical information of design and maintenance specialists related to the equipment and equipment of nuclear power plants by FAEA (failure mode effect analysis) or FME / CA (failure mode effect / criticality analysis) method, An example in which the present invention is applied to a thoughtful search method from an information bank that has been input to a database management device of a computer and made into a database will be described as an embodiment.

Also, this embodiment is based on the RCM (Reliability Centered Matrix).
(intenance: reliability-oriented maintenance), which is equivalent to information analysis processing for planning and deciding measures for equipment maintenance work. Conventionally, humans analyze data on data sheets and other papers.
The process under consideration corresponds to a computerized example. For this reason, it is possible to contribute to the rapid and sophisticated improvement of the conventional RCM information analysis processing.

In the information analysis for RCM implementation, the knowledge and technical information of the design and maintenance specialists are first analyzed by FMEA (failure mode influence analysis) or FME / CA (failure mode influence / criticality analysis). , As a data sheet as shown in FIG.

FIG. 5 shows an example of a technique for extracting knowledge and technical information at the system-device level, and FIG. 6 shows an example of a technique for extracting knowledge and technical information at the device-component level.

In the data sheet shown in Fig. 5, the equipment information such as "1. equipment identification number", "2. equipment name", "3. affiliated system name", etc., and "4. creation department, person in charge, creation date""
"5. Review department, person in charge, creation date", "6. Approval department, person in charge, creation date", "7. Revision department, person in charge, creation date",
Such as information management data, "8. Item (general name of equipment, equipment, parts, etc.) outline" as the data describing the function and use of the equipment, and "9. Organize and describe data such as drawing numbers. The FMEA or FME / CA data for the equipment is “10.FMEA”, which is the FMEA data of the entire equipment, for example, 1) importance, 2) margin for countermeasures (in the event of failure or failure), 3) design Data such as reliability and 4) maintainability are evaluated and described by data such as words or rank symbols identified by a rank setting table.

Further, as “11. Operation time”, the operation classification of the equipment, for example, is coded by a predetermined classification symbol such as during operation, and “12. Function (and required performance), for example,
In the case of a pump, an outline of functions, such as a function of performing a predetermined design or operation specified value function by two pumps, is described as “water supply function”. Further, a state in which the function is lost or deteriorated is described as “13. Functional failure (and related components)”.

In each of “12. Function” and “13. Functional failure”, an identification number is described so that the relation of data can be identified. In “13. Functional failure”, abnormal conditions of related units (components of equipment), for example, energy conversion system failure, etc.
The components related to the functional failure are described.

In addition, the following FMEA or FME / CA information is arranged and described for each identification number of 12. and 13. That is, for each function-function failure.

In the item of “14. Number of achievements”, the number of occurrences or assumed failure rate is described. Furthermore, as the "frequency", the frequency of occurrence is determined by a rank table that is determined in advance.
For example, "D; the probability of occurrence is extremely low"
Describe the rank symbol "D".

In the item of “15. Failure detection”, target data for detecting occurrence or sign of occurrence of a functional failure is specified as “1) plant data”. As “2) detecting means”, means for detecting the data of 1) is described.
Further, as "3) ability, 1) and 2), the ability to detect a functional failure is identified from a predetermined rank table, and a rank symbol is described.

In the item of “16. Failure effect”, the influence propagation process when a functional failure occurs will be described as “1) Influence process”. Further, as “2) margin for countermeasures”, a time margin for preventing and repairing the influence of a functional failure and a method of treatment are described and described. Further, as “3) influence”, a rank table of influence categories (for example, safety, drivability, etc.) of a predetermined functional failure is identified, and a rank symbol is described. As "4) degree", a rank symbol is described by identifying from a rank table of predetermined degree of functional failure influence (for example, automatic scrum, manual stop, etc.). Further, “5) margin” is a margin for countermeasures for a predetermined functional failure (for example, “D; there is no time limit”).
Identify from the rank table and enter the rank symbol.

In the item of “17. Repair conditions”, actions and conditions necessary for repair of functional failure are described, and “1) Time” is identified from the rank table of the repair time of functional failure that is determined in advance, and the rank symbol is used. Is described. Furthermore, "2) time"
Is identified from a predetermined rank list of functional repairable time and a rank symbol is described.

As “18. Current preventive measures and improvement plans”, their identification numbers and “1) Type” (eg, design improvement; DE, time-based maintenance; TD, condition monitoring maintenance; CD, etc.) Describe the content. What is not implemented at the moment,
Specify that this is a “draft”. As for “2) Capability”, for each preventive response and improvement plan, the functional failure prevention capability when it is implemented is identified from a predetermined rank table for functional failure preventive response / improvement capability, and the rank symbol is used. Is described.

In the item of “19. Importance”, the product or sum of the rank values given in advance to the various rank symbols described above is used in advance to determine the overall occurrence of functional failure, maintainability, impact, repairability, etc. A corresponding rank symbol is described from a rank table of importance determined for identifying as a result of the evaluation.

Finally, the data sheet (FMEA) shown in Fig. 6, which extracts and installs equipment and component level knowledge and technical information described below.
Data) is described. "Function-Functional failure" information with "Yes" is described in "13. Functional failure (related parts)"
-Component level FM for each related part extracted in item
EA is implemented.

In the data sheet shown in Fig. 6, equipment information such as "1. equipment identification number", "2. equipment name", "3. affiliated system name", etc., and "4. creation department, person in charge, creation date""
Information management data such as "5. Review department, person in charge, creation date", "6. Approval department, person in charge, creation date" and "7. Revision department, person in charge, creation date" are described. The data so far is the same as the data sheet shown in FIG.

Next, “13. Functional failure (and its related parts)” in the data sheet shown in FIG. 5 is described again, and the related parts described in FIG. 5 are further described as “20. Related parts (functions)”. I do. Each of “13. Functional failure” and “20.
An identification number is described so that the association of data can be identified.

In “20. Related Parts”, a functional overview of the part is also described.

In addition, the following FMEA or FME / CA information is organized for each identification number of 13. and 20.
Describe.

In the item “21. Failure mode”, “1. (Failure mode) phenomenon” and “2) (Failure mode) cause” of “20. Assign an identification number to the data. Hereinafter, the FMEA or FME / CA information is determined for each identification number by a word or a symbol and described. In the item of "21. Failure mode",
As “3) progress”, the form of the temporal change of the failure mode is identified from a predetermined rank table relating to the failure mode progress form, and the rank symbol is described. In addition, as "4)," the degree and speed of progress in 3) are identified from a predetermined rank table relating to the degree of progress of the failure mode, and the rank symbol is described.

In the item of “22. Results”, the frequency, degree and number of occurrences of the failure mode are described. Furthermore, as "frequency",
The frequency of occurrence is evaluated, for example, as “D; the possibility of occurrence is extremely low” by a predetermined rank table, and the rank symbol “D” is described.

In the item of “23. Cause detecting means”, means for detecting occurrence of a failure mode, target data, and the like are described. Also,
As “capability”, the capability of the detection means is identified from a predetermined rank table, and a rank symbol is described.

In the item “24. Effects of equipment failure”, “1) Effects of failure”
In the following, a description will be given of a process of transmitting an influence on a device when a failure mode occurs. Also, "2) Sign data detection"
The following describes device data affected by the occurrence of a failure mode and its detection means. Furthermore, as “3) influence”, the influence classification of the equipment failure is identified from a predetermined rank table, and the rank symbol is described. As “4) progress”, the influence progress mode of the device failure is identified from a predetermined rank table regarding the progress mode of the functional failure, and the rank symbol is described. Further, as "5) degree", the degree of progress of the equipment failure is identified from a predetermined rank table relating to the degree of progress of the equipment failure, and a rank symbol is described. As “6) detection”, the capability of the detection means is identified from a predetermined rank table, and a rank symbol is described.

In the item of "25. Repair conditions", actions and conditions necessary for repairing the functional failure are described, and "1) Margin" is identified from the rank table of the margin for countermeasures for the functional failure, and the rank symbol is used. Is described. “2) Time” is identified from a rank table of repair time for a predetermined functional failure, and a rank symbol is described. Furthermore, “3) Timing”
Is identified from a predetermined rank list of functional repairable time and a rank symbol is described.

As “26. Current preventive measures and improvement plans”, their identification numbers, and “1) Type” (eg, design improvement; DE, time-based maintenance; TD, condition monitoring maintenance; CD, etc.) Describe the content. What is not implemented at the moment,
Specify that this is a “draft”. As for “2) Capability”, for each preventive response and improvement plan, the functional failure prevention capability when it is implemented is identified from a predetermined rank table for functional failure preventive response / improvement capability, and the rank symbol is used. Is described.

In addition, if necessary, in the item of “19. Importance”, the value of the product or sum of the rank values given in advance to the various rank symbols described above can be used in advance to indicate the occurrence of functional failure, integrity, impact, and repair. A corresponding rank symbol is described from a rank table of importance determined for identification as an overall evaluation result such as gender. Other items such as words and rank information are added as necessary.

The information necessary for the RCM analysis as described above can be converted into a database by applying a computer. If there are multiple plants, create a database together with necessary information such as plant names.

Further, in the present embodiment, these pieces of information are converted by a computer using an information retrieval apparatus according to the present invention shown in FIG.
Information input / output display device 4, for example, a mouse on a CRT screen;
Using the information input device 5 such as a keyboard, the input can be easily made on the FMEA and FME / CA information input screen as shown in FIG. To input a rank symbol such as "frequency" or "ability", a sub-screen of the rank table is displayed on the information input / output display device 4 by pointing a data item portion on the screen with a mouse. Information can be input simply by pointing the symbol portion again with the mouse.

FIG. 8 shows an information input / output display device 4, for example, an information input device 5 such as a mouse and a keyboard on a CRT screen.
7 illustrates an embodiment for creating a logic tree.
FIG. 9 shows a completed example of a logic tree created using the thoughtful search function.

Hereinafter, with reference to FIG. 8 and FIG. 9, a procedure of creating a logic tree by the thoughtful search function and searching for information using the logic tree will be described.

(1) Select a thoughtful search function from the main menu of the system.

By this operation, a screen for creating a logic tree as shown in FIG. 8 is displayed.

(2) On this screen, first, basic information such as a creator and a creating department is input and set by using an information input device such as a mouse and a keyboard.

(3) Next, the data to be searched is searched by the database, and the data is set and stored in the information processing / storage device.

In this embodiment, from the RCM analysis information database, the search condition setting items in the upper part of FIG. By instructing, search conditions such as a plant name and a device name are set.

Then, a search is executed by instructing a “search” button among the operation buttons in the upper part of the screen with a mouse.

(4) Next, a logic tree is created on the screen using a mouse and a keyboard.

The sub-screen (logic,
From the menu for tree creation), point the "Select" button with the mouse using the operation buttons at the bottom of the sub screen, and then, 1) conditional branch (diamond), 2) branch condition (elongated oval), 3) output in the screen Use the mouse to specify the required figure from the figures such as conditions (rectangle), 4) comment (oval), and further specify the "move" button with the mouse using the operation buttons at the bottom of the sub screen, and specify the figure with the mouse While moving the mouse, a necessary figure is moved and set to an appropriate place on the screen by a method such as moving the mouse.

A logic tree is created in a similar manner.

The “erase” button is selected when erasing a figure once set.

The “display adjustment” button is selected when a logic tree created as a collapsed shape is automatically corrected to an optimal arrangement.

The “connection” button is selected when connecting a set figure with a line between figures.

The “Create” button is selected when creating and adding a new target figure.

And other operation functions, which can be added as needed.

It is assumed that the size and shape of the figure can be freely changed by operating the mouse.

Next, from the sub-screen (condition setting menu) that is automatically displayed, the necessary items,
It is set by designating a search condition with a mouse.

Similarly, in 3) output condition (rectangle), a data item to be output is designated and set by a mouse on a sub-screen (condition setting menu) in each figure.

Identification items and the like are input and set from a keyboard or the like.

By performing the above operation, a setting example of the search condition example of the thoughtful search (logic tree) from the RCM analysis information database is created as shown in FIG.

In the case of a logic tree that does not fit on one screen, it can be expanded by pointing the "next page / previous page" at the top of the screen with the mouse.

(5) If necessary, register the created logic tree in the information processing / storage device and execute a thoughtful search.

To register a logic tree, click “Register” at the top of the screen.
It is executed by pointing a button with the mouse.

The thoughtful search is executed by instructing a “search” button at the top of the screen with a mouse.

The logic tree that has already been created and registered is
Indicate the processing mode at the top of the screen with the mouse and select “New →
Update ”(if creating a new one, leave it as is)
From the keyboard to the "basic pattern" part,
Enter the name of the tree. By instructing the "search" button with the mouse, the registered logic tree is searched by the information processing / recording device, and is displayed on the screen together with the accompanying information. If you have more than one of these suggestions,
“Sheet” and “Number of candidates” are displayed, and “Next candidate /
By instructing the "pre-candidate" button with a mouse, a necessary logic tree can be selected and displayed.

The “delete all” button deletes the displayed logic tree from the information processing / storage device at once. The "delete" button simply deletes the display from the screen.

In the setting example of the search condition shown in FIG. 9, the “failure influence” is obtained from the data registered in the RRCM analysis information database by the FMEA-based information extraction and management as described in (1).
By combining the five types of data items of “failure detection”, “repair time”, “repair margin”, and “occurrence frequency”, the priority of the failure mode requiring design improvement among the failure modes of the equipment is set to “priority ( A logic tree is shown that classifies into four types of decision levels of "large", "priority (medium)", "priority (small)", and "no improvement required", that is, classification items and outputs search results. I have.

A, B, C, and D in the figure indicate the rank symbols of the respective data items, and the search is executed according to the branch of the logic tree in accordance with the rank symbol data of each failure mode.

For example, “Failure mode A” has a rank symbol of “A”, a rank symbol of “failure detection” is “C”, a rank symbol of “repair time” is “B”, and a rank symbol of “repair margin”. If the rank symbol is “C” and the rank symbol of “occurrence frequency” is “C”, this failure mode is classified into the “priority (medium)” identification item according to the logic tree.

The results searched by the method and function of the above (2) are displayed on an information input / output display (CRT) as shown in FIG. 10, for example. Is output.

FIG. 10 shows an example in which information corresponding to the identification item “priority (medium)” is selectively summarized in a table format from information retrieved according to the logic tree shown in FIG. 9 and output. I have.

As described above, the output of the search result can output the designated information or all the information collectively.

In other words, the user's thinking process can be logically expressed and visualized by a logic tree for the initial purpose of “evaluation of failure mode priority requiring design improvement”, which is a decision-making problem. A highly advanced search can be made possible as compared with the conventional search method. Furthermore,
This makes it possible to simultaneously perform decision-related identification, which cannot be performed by a conventional search, to effectively support the user's decision-making. Further, when the search according to the present embodiment is performed by the conventional search method, it is necessary to repeat the search by the number of branches of the logic tree (9 times in the example of FIG. 9). According to the method of the present invention, information can be extracted by one search, which can greatly contribute to efficient search.

〔The invention's effect〕

As described above, the present invention is applicable to a case where there are a plurality of judgment conditions related to search information from an information bank such as a database management device realized by a computer, etc. And the like are expressed as search conditions in the form of a linked logic tree or the like based on individual judgment conditions, and are set and searched at each final branch point or in the middle of a branch. The target information is arranged in a tabular format and the like according to the purpose of judgment and identification of each branch point together with the preset search output target information, and the information is output, so that the user can simply output the information from the information bank. In addition to extraction, information in the form of supporting user decision-making,
Extraction can be performed according to the user's thinking system, and the range of use of the conventional database management system can be expanded.

In addition, as shown in the present embodiment, the present invention provides design and maintenance technology for equipment and equipment of a large plant such as a nuclear power plant.
When applied to a knowledge information database, it not only establishes an efficient management system for technical and knowledge information, but also evaluates the importance of design improvement points, examines maintenance content improvement plans, It is possible to provide useful information that can be useful for assisting in planning a maintenance plan of parts. As a result, failures can be prevented beforehand by implementing appropriate maintenance measures, and effective information that can greatly contribute to the improvement of the reliability and reliability of plant equipment can be provided easily and quickly. It has the effect.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a configuration of an information search device according to the present invention, FIG. 2 is an explanatory diagram showing an example of an information search process using a search method according to the present invention, and FIG. FIG. 4 is a diagram showing an example of an information output display in the middle of a search searched using the present invention, FIG. 4 is a diagram showing an example of an output information display of a search result searched by the present invention, and FIG. Data sheet diagram showing an example of a method for extracting knowledge technology information at the system-device level,
FIG. 6 is a data sheet diagram showing an example of a technique for extracting knowledge and technical information at a device-part level using the present invention, FIG. 7 is a diagram showing an example of a screen of an information input / display device according to the present invention, and FIG. The figure shows logic using the thoughtful search function according to the present invention.
FIG. 9 is a diagram showing a tree creation method, FIG. 9 is a diagram showing an example of creation of a logic tree according to the present invention, FIG. 10 is a diagram showing an information output display example of a result of information retrieval according to the present invention, and FIG. FIG. 12 is a diagram showing an example of database contents in a format, and FIG. 12 is a diagram showing an example of information output by a conventional search method. 1 ... Computer, 2 ... Information processing / recording device, 3 ...
Database management device, 4 ... information input / output display device, 5
... an information input device, 6 ... an information output device.

Claims (2)

(57) [Claims] A step of expressing and setting a request such as a searcher's information search purpose and a decision making procedure as a search condition in the form of a logic tree in which branches based on individual judgment conditions are developed in stages and connected. Arranging the target information retrieved at each final branch point or in the middle of the logic tree according to the purpose of discriminating and identifying each branch point, and outputting the information, and outputting the information from the database management apparatus. . 2. A database management device in which search information is collected and stored; an information calculation processing / recording device connected to the database management device for performing arithmetic processing and storing the search information; An information input device connected to a processing / storage device for performing information input operation; and an information input / output display device connected to the information calculation processing / storage device for displaying an information input / output state, The processing / storage device expresses the search condition in the form of a logic tree, organizes the target information searched at each final branch point or the middle of the branch of the logic tree according to the purpose of determining and identifying each branch point, and outputs the information. An information retrieval apparatus characterized by having a thoughtful retrieval function.