JP2001273164A - System for diagnosing failure of electronic equipment and electronic component unit and failure analyzing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a failure diagnosis system and a failure analyzing method for electronic equipment and an electronic component unit, which guide an inexperienced person such as a part-timer in repair parts to enable the person to easily repair electronic equipment and an electronic component unit without depending on a professional engineer. SOLUTION: A failure phenomenon sentence of a natural sentence is developed into specified words used as terms that technically express a failure phenomenon in accordance with each word obtained by decomposing the failure phenomenon sentence of the natural sentence into nouns and verbs respectively, technical information showing the contents of the failure phenomenon is obtained on the basis of the specified words, and the failure position information and the failure electronic component information of the electronic component constituting the electronic component unit are obtained on the basis of the technical information and information specifying the electronic component unit with a failure.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a failure diagnosis system and a failure analysis method for an electronic device and an electronic component unit, and more particularly to a method for guiding a repair part to an amateur such as a part-timer without using a specialized engineer. The present invention relates to a failure diagnosis system and a failure analysis method for an electronic device and an electronic component unit that can easily repair a component unit.

[0002]

2. Description of the Related Art Conventionally, repair of an electronic device is performed by a technical engineer who has received specialized education in each technical field or an engineer who has a lot of experience in the field. In particular, in repairing electronic devices such as computers, a specialized repair technician has accumulated know-how by himself, performed a failure diagnosis based on the accumulated know-how, identified a defective component, and replaced it. Computer Telephony Integration placed at corporate inquiry centers as this type of conventional technology
(CTI). This is to estimate the cause from the failure or symptom event input as a diagnostic expert system. Usually, for diagnosis, there is a specialized intermediary between the database and the intermediary. This is a system that responds to inquiries.

[0003]

In the case of repairing an electronic component unit, a professional engineer or a basic knowledge of repair is absolutely necessary. However, it takes time to train specialized engineers or repair staff who have received specialized education in each technical field, and the number of them is limited. Also, few engineers have much experience in the field of repair technology. In addition, the cost required for maintenance and repair also increases. On the one hand, many low-cost electronic devices have been developed,
The number of repairs and maintenance is on the rise, and is on the rise. SUMMARY OF THE INVENTION An object of the present invention is to solve such a problem of the prior art, and to guide a repair part to an amateur such as a part-timer to repair an electronic device and an electronic part unit without a specialist. It is an object of the present invention to provide a failure diagnosis system and a failure analysis method for an electronic device and an electronic component unit which can be easily performed.

[0004]

A feature of the electronic device and the electronic component unit failure diagnosis system according to the present invention for achieving the above object is that a failure sentence of a natural sentence is converted into a noun and a verb, respectively. First storage means for storing a specific word used as a term that expresses a failure phenomenon professionally corresponding to each word obtained by disassembly; and a professional memory indicating the content of the failure phenomenon corresponding to the specific word. Second storage means for storing important information, and comparing the failure position information and the failure electronic component information of the electronic components constituting the electronic component unit in correspondence with the specialized information and the information of the electronic component unit. A third storage unit, an input unit for inputting information for specifying a faulty electronic component unit, and a first process of decomposing a failure phenomenon sentence of a natural sentence of the input failure into words of a noun and a verb, respectively. Note First selecting means for searching a storage means and selecting a specific word suitable according to the decomposed word;
A second selection unit that searches the second storage unit based on the specific word selected by the first selection unit and selects specialized information indicating the content of the failure phenomenon; and the second selection unit. The third storage means is searched based on the specialized information indicating the content of the failure phenomenon obtained by the above and the information for specifying the input electronic component unit, and the failure position information and the failed electronic component for the electronic component unit are retrieved. Information and third choice
And output means for outputting fault position information and fault electronic component information obtained by the third selecting means.

A feature of the failure analysis method of the present invention is that
Breaking down the failure sentence of a natural sentence into nouns and verbs, each word is obtained by decomposing it into a specific word that is used as a term that expresses a failure phenomenon corresponding to each word obtained. Obtains specialized information indicating the content of the failure phenomenon based on the failure information, and based on the specialized information and the information specifying the failed electronic component unit, failure position information and failure of the electronic components constituting the electronic component unit. The electronic component information is obtained.

[0006]

As described above, according to the present invention, specialized information indicating the contents of a failure phenomenon is obtained from a failure phenomenon sentence of a natural sentence, and the information is used together with an electronic device or an electronic component unit having a failure. Since the failure position information and the failure electronic component information of the electronic component are obtained based on the information specifying the electronic component, the failure position information and the failure electronic component information of the electronic component can be easily obtained even by an amateur. As a result, it is possible to easily repair the electronic device or the electronic component unit by guiding the repair part to an amateur such as a part-timer without the need of a specialized technician. In particular, it is possible to provide a means for a non-expert to systematically solve a diagnostic technique that requires expertise that cannot be solved only by skill.

[0007] The present invention, particularly in the repair of an electronic device such as a computer, can eliminate the specialty in this field as much as possible, and can provide a means enabling non-specialists to repair the device.

[0008]

FIG. 1 is a block diagram of one embodiment of a failure diagnosis system for an electronic component unit according to the present invention, FIG. 2 is a flowchart of a repair content analysis process, and FIG. FIG. 4 is an explanatory diagram of a repair dictionary database, FIG. 5 is an explanatory diagram of a display screen of a search result of the repair dictionary database, and FIG. 6 is a flowchart of a search process for obtaining a similar repair word and a registered phenomenon sentence. FIG. 7 is an explanatory view of a repair version table, FIG. 7 is an explanatory view of a data input screen, and FIG.
It is explanatory drawing of the output screen which outputs repair information. 1, a failure diagnosis system 1 for an electronic component unit includes an input device 2 such as a keyboard for inputting repair information for each item, a printer 3, and a failure diagnosis terminal device 4.

The failure diagnosis terminal device 4 is provided with a similar repair word dictionary database 48, a repair word failure phenomenon conversion dictionary database 49, and a repair dictionary database 50, and the input phenomenon information (one described in a natural sentence). ), A similar repair word dictionary database 48 is searched to extract a specific word (hereinafter referred to as a repair word) used as a term that expresses a failure phenomenon from a word of a natural sentence, and further repairs from the repair word. The word failure phenomenon conversion dictionary database 49 is searched, and the phenomenon information of the natural sentence is converted into a specific specialized failure phenomenon sentence for dictionary search. Then, the storage corresponding to the electronic component unit to be repaired in the repair dictionary database 50 based on the specialized failure phenomenon statement and the item data (specific data for specifying the electronic component unit to be repaired) input for the repaired electronic component unit. The area is searched, and the repaired part and the repaired part are stored in the printer 3.
Is displayed on the display 45.

In the failure diagnosis terminal device 4, an MPU 41, a memory 42, an interface 43, a communication control circuit 44, a display 45, and the like are mutually connected via a bus 46. An external storage device 47 such as a hard disk, the input device 2 and the printer 3 are connected to the interface 43. The external storage device 47 has a similar repair word dictionary database 48, a repair word failure phenomenon conversion dictionary database 49, and A repair dictionary database 50 is stored. The memory 42 stores a word decomposition AI failure phenomenon conversion program 42a, a repair information search program 42b, a repair information format output program 42c, and the like as processing programs to be executed by the MPU 41, and has a work area 42d. The communication control circuit 44 is connected to the line 5
And receives, via a line 5, various item data input to the input device 2 and the content of a failure phenomenon. In the following description, the item data input via the line 5 and the content of the failure phenomenon are the same as in the case of the input device 2 and will not be described.

A word decomposition AI fault phenomenon conversion program 42
“a” is called from the main program in response to the end of the input of the item data specifying the target and the data of the failure phenomenon information for the repair electronic component unit, and is executed by the MPU 11. When this is executed, MPU1
1 is a method of decomposing a phenomenon sentence written in a natural sentence about an inputted failure into words of a noun and a verb, searching a similar repair word dictionary database 48, and selecting a repair word from words of the natural sentence, Based on a plurality of AND conditions of each word of the selected repair word and OR conditions of each word, the repair word failure phenomenon conversion dictionary database 49 is searched, and specific technical failures registered in the descending order of the hit rate are registered. A plurality of phenomenon statements (failure phenomenon information) are extracted. Then, the repair information search program 42b is called. The registered professional failure phenomenon sentence is stored in the repair word failure phenomenon conversion dictionary database 49.
It is a keyword provided for searching.
The repair word is composed mainly of words (nouns and verbs) used in specialized failure phenomenon sentences.

When the repair information search program 42b is executed, the MPU 11 determines the corresponding electronic component unit in the repair dictionary database 50 from the extracted specialized failure phenomena statement and the item data specifying the electronic component unit. With reference to the area, data of the repair target component, its mounting position, and the failure frequency in the column where the specialized failure phenomenon statement matches are extracted. Further, the data of the parts to be repaired and their mounting positions are extracted for each failure occurrence frequency, and the repair information format output program 42c is called. When the repair information format output program 42c is executed, MP
U11 selects the parts having the highest frequency of failure occurrence in the order of the hit rate (priority order) of the failure phenomena shown in FIG. 5 for the obtained repair data, and shapes the repair contents into a predetermined format that is easy to understand. It is displayed on the display and processed for printing on the printer. When the next occurrence frequency is designated by the input data, the faulty component having the next occurrence frequency is selected in the order of the hit ratio (priority) of the failure phenomenon sentence.

Next, the flow of the entire process of the fault diagnosis will be described with reference to FIG. In FIG. 2, first, a repair number and a customer name are input (step 101), and a symbol indicating the electronic component unit (for example, a board name PC) is used to specify which component the repair electronic component unit is.
B1) Enter the design lot number (step 102)
The content of the failure phenomenon is input as it is in a natural sentence expression (step 103). In this case, the failure phenomenon sentence written on the paper may be read by a reader, a scanner, or the like. Each input data is stored in the work area (step 104). Next, the word decomposition AI fault phenomena conversion program 42a is called and executed by the MPU 11, and the content sentence of the fault phenomena is read from the work area 42d (step 105), and nouns and verbs are word-decomposed to extract nouns and verbs. (Step 10
6).

Then, the similar repair word dictionary database 48 is searched from the decomposed noun and verb, and a plurality of similar repair words in a professional sense are extracted by analogy or the like by AI processing (step 107). The repair word failure phenomenon conversion dictionary database 49 is obtained by using a plurality of AND conditions and a OR condition of each word from the extracted special repair words.
And extracts a plurality of registered specific technical failure phenomena sentences in descending order of the hit rate (step 10).
8). Since there are a plurality of repair words, a hit rate is obtained as a hit when each of the failure phenomena sentences is selected.

FIG. 3 is an explanatory diagram of a search process for obtaining this specialized failure phenomenon sentence. First, it is assumed that the phenomenon information D1 in the natural sentence describing the phenomenon of the failure input in step 103 is described as "the screen is out of focus when a blurry light comes out and the screen appears." I do. The words are decomposed into nouns and verbs, and the decomposed word group D2 is "screen, blurred light, emerge, focus, blur". The analogous word of the specialized repair obtained by searching the similar repair word dictionary database 48 from these words is “screen, back raster, out, focus, focus, bokeh, blur, blur” as a word group D3. And so on. Then, as a professionally observed repair phenomenon obtained by combining these words under the AND condition and the OR condition, D4 indicates "back raster is out of focus and the screen is discolored to green and the focus is blurred" as D4 in descending order of the hit rate. , "And" the screen is entirely blurred. "

Next, the area of the corresponding electronic component unit in the repair dictionary database 50 (see FIG. 4 and described later) is determined from the extracted specialized failure phenomenon statement and the item data specifying the electronic component unit. The data of the repair target component, its mounting position, and the failure frequency are extracted from the column where the specialized failure phenomenon statement matches with reference to the column (step 109). The similar repair word dictionary database 48 is used to extract similar repair words by AI analogy from words that are usually used by amateurs and that mean troubles or their symptoms, and have the same structure as the Japanese conversion program. The input word is simply a word obtained by decomposing a natural sentence, and the word to be selected is merely a similar repair word. The failure phenomenon conversion dictionary database 49 is a table in which a phenomenon sentence is described using similar repair words and their combinations as search keys, and stores a high probability phenomenon from the AND condition and OR condition of similar repair words. It is a dictionary that is extracted in order.

FIG. 5 shows the result of searching the failure phenomenon conversion dictionary database 49 on the display 45. In FIG. 5, a display column 401 is a column for displaying a repair number, and is the number input in step 101. The display column 402 is a column for the design lot number of the electronic component unit name (for example, board name PCB1) input in step 102. The reception history column 408 and the shipping history column 409 are display columns for old and new version information described later. The customer name field and the site phenomenon field 404 are the customer name input in step 101 and the content of the failure input in step 103 in a natural sentence. Here, for convenience of explanation, an example of a failure of the printer will be described below. The column 405 is the failure phenomenon conversion dictionary database 4
9 is a column in which a phenomenon sentence obtained by searching for No. 9 is displayed. This is displayed according to the rank of the hit rate. The column 406a indicates the priority, the column 406b is a phenomenon column that has hit, and the column 406c displays the hit rate as a weight.

FIG. 4 is an explanatory diagram of a repair dictionary database 50 for specifying a failed part and a repair part from the extracted failure phenomenon sentence. A large number of repair dictionary databases 50 are provided corresponding to target electronic component unit identification codes. For example, if the identification code is a circuit board, a PC
The table of BXXXXXX is 50a. Table 50a
Is described in the failure phenomenon column 51, and is searched using this as a key. Then, in the next component name column 52, the component names of the defective components are sequentially described in descending order of the failure probability. The mounting position column 53 describes the mounting position of the failed component. The occurrence frequency column 54 describes a failure occurrence rate, and the table is managed in descending order of the failure rate.
The failure rate is calculated from the past failure history and updated. The last column 55 is a replacement parts list column of the latest version, in which data of replacement parts and their mounting positions are stored.

Now, step 11 following step 109
In the case of 0, it is determined whether or not to input a reproduction phenomenon in a failed electronic device or the like. If YES, the reproduction phenomenon flag is set, and the process returns to step 103, where the reproduction phenomenon is similarly input in natural language, and the processing from step 104 onward is performed. Then, similarly, in step 109, data on the repair target component, its mounting position, and failure frequency are extracted. This is separately treated as a reproduction phenomenon as described later with reference to the reproduction phenomenon flag, and is displayed on the display 45.

In the next step 111, referring to the repair version table of FIG. 6, the old version information of the electronic component unit name (for example, the board name PCB1) and the latest version information when the component is replaced with the latest version component. Is obtained, and it is determined whether or not the part requires adjustment based on the design lot number input in step 102 (step 11).
2) If adjustment is required, the result is YES, and data is obtained from the design lot number with reference to the adjustment content storage table (not shown) (step 113). If no adjustment is required, the determination is NO, and the process proceeds to step 114.

FIG. 6 shows a repair version table, which is the latest version when the current version A is replaced with the current version A using the electronic component unit name (eg, board name PCB1) input in step 102 as a key. D is stored. Next, the repair information format output program 42c is called and executed by the MPU 11. A process of outputting a list of repair information in which repair phenomenon statements are selected in order of priority is performed on the printer 3,
The repair contents are displayed on the display 45 in the order of the repair priority from the highest priority as shown in FIG. 7 (step 114). The contents output to the printer 3 are substantially the same as the contents displayed on the display 45 and are arranged in the order of priority, so that the description is omitted. As described above, when a component having the next occurrence frequency is designated, a faulty component having the next occurrence frequency is selected. I just need. In step 114, the next failed electronic component is selected and displayed.

FIG. 7 shows output data of the repair information. In FIG. 7, a display column 801 is a column for displaying a repair number, which is the number input in step 101. The display column 802 is a content in which the customer expresses the failure in a natural sentence. A display column 803 is a column in which a failure phenomenon sentence obtained by searching the failure phenomenon conversion dictionary database 49 for a reproduction phenomenon is displayed. A failure phenomenon sentence obtained by searching the failure phenomenon conversion dictionary database 49 with a sentence is displayed as a reproduction phenomenon. The display column 807 displays the old version information A as the reception history column, and the display column 808 displays the latest version information D as the shipment history column.

The display column 804 is for the repair dictionary database 5
In the area in which the contents of "0" are displayed, the faulty component and its mounting position are displayed in a column 805 as the repair content, and the remodeling content of the latest version read from the column 55 of the repair dictionary database 50 is displayed in a column 806. You. Further, if adjustment is necessary in the determination of step 112, the details of the adjustment are displayed thereafter. The last inspection content column is a column for displaying the content when an inspection is required along with the adjustment content stored in the adjustment content storage table.

As described above, in the embodiment, the failure phenomena sentence of the natural sentence input is not limited to the input by the input means such as the keyboard, but may be the data transmitted through the line. Alternatively, a sentence read in a predetermined format by an image reader or the like may be used. Further, in the embodiment, the example of the electronic component unit is described. However, it goes without saying that the present invention can be similarly applied to an electronic device including a plurality of electronic components or electronic component units. Further, in the embodiment, a specific word used as a term that expresses a failure phenomenon professionally is stored corresponding to each word obtained by decomposing a failure phenomenon sentence of a natural sentence into each word of a noun and a verb. A first database, a second database for storing specialized information indicating the content of the failure phenomenon corresponding to the specific word, and an electronic component unit corresponding to the specialized information and the information of the electronic component unit And a third database for comparing and storing the failure position information of the electronic component and the failure electronic component information, which is used as a single database, each data is stored as one database. Is also good. Each data acquisition can be obtained by AI processing from one database. Of course, these databases may be general storage means.

[0025]

As can be understood from the above description, according to the present invention, specialized information indicating the content of a failure phenomenon is obtained from a failure phenomenon sentence of a natural sentence, and the electronic information or the electronic device having a failure is obtained. Since the failure position information and the failure electronic component information of the electronic component are obtained based on the information for specifying the electronic component unit, even a layman can easily obtain the failure position information and the failure electronic component information of the electronic component. . As a result, it is possible to easily repair the electronic device or the electronic component unit by guiding the repair part to an amateur such as a part-timer without the need of a specialized technician. In particular, it is possible to provide a means for a non-expert to systematically solve a diagnostic technique that requires expertise that cannot be solved only by skill.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a failure diagnosis system for an electronic component unit according to the present invention.

FIG. 2 is a flowchart of a repair content analysis process.

FIG. 3 is a flowchart of a search process until a specialized similar repair word and a registered phenomenon sentence are obtained from the decomposed words.

FIG. 4 is an explanatory diagram of a repair dictionary database.

FIG. 5 is an explanatory diagram of a display screen of a search result of a repair dictionary database.

FIG. 6 is an explanatory diagram of a repair version table.

FIG. 7 is an explanatory diagram of an output screen for outputting repair information.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Failure diagnosis system of electronic component unit, 2 ... Input device, 3 ... Printer, 4 ... Failure diagnosis terminal device, 5 ... Line,
41: MPU, 42: memory, 42a: word decomposition AI failure phenomenon conversion program, 42b: repair information search program, 42c: repair information format output program, 4
2d: Work area, 43: Interface, 44: Communication control circuit, 45: Display, 46: Bus, 47: External storage device, 48: Similar repair word dictionary database, 49:
Repair word failure phenomenon conversion dictionary database, 50 ... Repair dictionary database.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification code FI Theme coat ゛ (Reference) G06F 19/00 130 G06F 19/00 130 (72) Inventor Sadao Soga 456 Nakaicho, Ashigakami-gun, Kanagawa Pref. Hitachi Information Technology F term (reference) 5B048 AA20 DD11 FF02 5B049 AA06 BB07 CC00 DD01 DD05 EE05 FF02 FF03 GG04 GG07 5B075 ND03 NK06 NK34 NK37 NK39 NK54 PP22 PP25 PQ02 PQ03 PQ32 PQ74 PR06 PR08

Claims (7)

[Claims] 1. A method for storing a specific word used as a term for expressing a failure phenomenon professionally corresponding to each word obtained by decomposing a failure phenomenon sentence of a natural sentence into words of a noun and a verb. 1 storage means, second storage means for storing specialized information indicating the content of the failure phenomenon corresponding to the specific word, and the electronic device corresponding to the specialized information and the information of the electronic component unit. Third storage means for comparing and storing the failure position information and the failure electronic component information of the electronic components constituting the component unit; and input means for inputting information for identifying the failed electronic component unit. The first sentence is selected by decomposing a failure phenomenon sentence of a natural sentence relating to the failure into words of a noun and a verb, searching the first storage means, and selecting the specific word that fits according to the decomposed word. Means of selection, A second selection unit that searches the second storage unit based on the specific word selected by the first selection unit and selects specialized information indicating the content of the failure phenomenon; Searching the third storage means based on the specialized information indicating the content of the failure phenomenon obtained by the selecting means and the information for specifying the input electronic component unit, A third selection unit that selects the failure position information and the failure electronic component information; and an output unit that outputs the failure position information and the failure electronic component information obtained by the third selection unit. A failure diagnosis system for an electronic component unit. 2. The information for specifying the electronic component unit includes:
A symbol indicating the electronic component unit and a design lot number thereof, wherein the second selecting means manages a hit rate at which specialized information indicating the content of the failure phenomenon is obtained in accordance with the specific word; The specialized information is selected by assigning priorities in descending order of the hit rate, and the third selecting means generates identification information for identifying the electronic component unit from the symbol and the design lot number. 2. The failure diagnosis system for an electronic component unit according to claim 1, wherein said failure position information and said failure electronic component information are selected based on leverage identification information and said specialized information having a high priority. 3. A specific word used as a term specially representing a failure phenomenon corresponding to each word obtained by decomposing a failure phenomenon sentence of a natural sentence into words of a noun and a verb. 1 storage means, 2nd storage means for storing specialized information indicating the content of the failure phenomenon corresponding to the specific word, and the electronic device corresponding to the specialized information and information of the electronic device. Storage means for comparing and storing the fault position information and the fault electronic component information of the electronic components constituting the electronic device, input means for inputting information for identifying the faulty electronic device, First selecting means for decomposing the failure sentence sentence of the natural sentence into words of noun and verb, searching the first storage means and selecting the specific word suitable according to the decomposed word And the first selecting means A second selection unit that searches the second storage unit based on the selected specific word and selects specialized information indicating the content of the failure phenomenon; and a second selection unit that is obtained by the second selection unit. The third storage unit is searched based on the specialized information indicating the content of the failure phenomenon and the information specifying the input electronic device, and the failure position information and the failed electronic component of the electronic device are searched. 3. A failure diagnosis of an electronic device, comprising: third selection means for selecting information; and output means for outputting the failure position information and the failure electronic component information obtained by the third selection means. system. 4. The information for specifying the electronic device is a symbol indicating the electronic device, and the second selecting means obtains specialized information indicating the content of a failure phenomenon corresponding to the specific word. The third selection unit is configured to manage the hit ratios assigned to the electronic devices and assign the priority order in descending order of the hit ratios to select the specialized information. 4. The failure diagnosis system for an electronic device according to claim 3, wherein the failure position information and the failure electronic component information are selected based on the generated identification information and the high-priority specialized information. 5. The method according to claim 1, wherein the failure sentence of the natural sentence is decomposed into words of a noun and a verb, and the word is expanded into a specific word used as a term to express the failure phenomenon in correspondence with each word obtained. The electronic information constituting the electronic component unit is obtained based on the specialized information and the information specifying the electronic component unit having a failure, by obtaining specialized information indicating the content of the failure phenomenon based on the specific word. A failure analysis method for an electronic component unit, comprising obtaining component failure position information and failed electronic component information. 6. The information for specifying the electronic component unit,
A symbol indicating the electronic component unit and its design lot number, and manages the hit rate of the information when specialized information indicating the content of the failure phenomenon is obtained corresponding to the specific word. The special information is selected by assigning a priority order in descending order of the number, identification information for identifying the electronic component unit is generated from the symbol and the design lot number, and the identification information and the high 6. The failure analysis method for an electronic component unit according to claim 5, wherein the failure position information and the failed electronic component information are obtained based on the information. 7. A natural word failure phenomenon sentence is decomposed into a noun and a verb, and each word is obtained by decomposing the same into a specific word that is used as a term that expresses a failure phenomenon in a professional manner. Obtaining specialized information indicating the content of a failure phenomenon based on the specific word, and failure of an electronic component constituting the electronic device based on the specialized information and information for identifying a failed electronic device. Failure analysis method for electronic device characterized by obtaining position information and failure electronic component information