WO2020071054A1

WO2020071054A1 - Fault factor priority presentation device

Info

Publication number: WO2020071054A1
Application number: PCT/JP2019/035153
Authority: WO
Inventors: 清水　勇喜
Original assignee: 株式会社日立製作所
Priority date: 2018-10-02
Filing date: 2019-09-06
Publication date: 2020-04-09
Also published as: JP2020057192A; JP7154922B2

Abstract

An objective of the present invention is to provide a fault factor priority presentation device with which priority for a fault factor which is included in a fault tree can be evaluated and presented, and the identification of the fault factor can be hastened. The present invention comprises: a causal relation database (112) which stores causal relations which represent chains of a plurality of elements for a product malfunction occurrence which include components and phenomena; an output device (106) which displays a fault tree which relates to the product malfunction occurrence, in which said fault tree branches out from a top event to basic events; and an inter-event causal probability computation unit (111) which computes the probability of a path from the top event to a basic event from the causal relations stored in the causal relation database (112). According to the present invention, the probability computed with the inter-event causal probability computation unit (111) is displayed in the fault tree.

Description

Failure factor priority presentation device

<< The present invention relates to a failure factor priority presenting apparatus in a failure tree.

Fault tree analysis (FTA, Fault Tree Analysis) is known as a technique for preventing product failure. The FTA is an analysis technique for systematically searching for the source of a failure by taking up a failure event of a product, sequentially identifying and developing the failure factors in a hierarchical manner.
This analysis result has a tree structure in which a product failure event is at the top and the failure factor is a lower hierarchy. The analysis result having the tree structure is called a “fault tree”. A failure event of a product to be analyzed is called a “top event” because it is located at the top of the failure tree. The fault tree is composed of a plurality of layers. The fault factor at the end of the fault tree is called an “end event” because it is located at the end of the fault tree. The terminal event represents the root cause of the top event.

For example, when analyzing the cause of a failure of a certain system, the failure event of the system is described as a top event, and then the failure of the subsystem which is the cause of the failure of the system is identified and described in the lower hierarchy of the top event. Subsequently, the failure of the component which is the cause of the subsystem failure is identified and described in the lower hierarchy of the subsystem failure factor. In this way, the root cause, that is, the failure factor up to the terminal event is identified.

FTA can be used in the case where the reliability of a product is created at the design stage, and in the case where a defect occurs in a product, the cause of which is investigated. When used at the design stage, reliability is set by setting failure events that do not want the product to occur at the top event, and taking measures to prevent the root event, which is the root cause, obtained as an analysis result from occurring. To improve. In the failure factor analysis after the failure occurs, the generated failure event is set as a top event, an FTA is performed, and it is confirmed whether or not the terminal event is actually a failure factor of the top event.

At this time, if there are many end events that have been identified, it is more efficient to assign priorities and consider the highest priority.

技術 As a technique for searching for a cause of a failure, for example, Patent Document 1 is proposed. In the technique described in Patent Document 1, the priority is calculated based on the number of occurrences of the failure factor and the freshness calculated based on the number of days elapsed from the latest occurrence date.

JP 2009-217457 A

(4) In Patent Document 1, a failure factor, that is, the number of occurrences of a terminal event is accumulated, and the priority is determined based on this. This is based on the idea that the higher the number of occurrences of the terminal event, the higher the probability of causing the top event.

However, since the failure is caused by a chain of multiple events, for example, "initial crack" → "crack propagation" → "breakage", not only the failure rate of the terminal event but also the path between the top event and the terminal event Needs to be considered. That is, in the above example, the probability between events is (probability that the cause of the top event “breakage” is “crack propagation”) × (probability that the cause of “crack propagation” is “initial crack”). Is required. The technique described in Patent Literature 1 has a problem that only the end event is considered, and the probability of a path between the top event and the end event is not considered.

In addition, the probabilities of all combinations between events such as (probability that the cause of the top event “breakage” is “crack propagation”) and (probability that the cause of “crack propagation” is “initial crack”) are prepared in advance. To do so takes time. Further, this probability varies depending on the surrounding environment and usage of the target product / part. Even if the same product / part is used in the nuclear field and in home appliances, the probability will be different. In other words, the probability varies depending on fields such as nuclear power and home appliances.

An object of the present invention is to provide a failure factor priority presentation device that can evaluate and present the priorities of failure factors included in a failure tree and can quickly identify the failure factors.

In order to achieve the above object, the present invention is characterized by a causal relationship database storing a causal relationship representing a chain of a plurality of elements including parts and phenomena with respect to occurrence of a product defect, and from a top event to a terminal event. An output device for branching toward and displaying a fault tree related to the occurrence of a product defect; and an event-to-event causal probability calculation for calculating a probability of a path from the top event to the terminal event from the causal relationship stored in the causal relationship database. And displaying the probability calculated by the inter-event causal probability calculation unit on the fault tree.

According to the present invention, since the priorities of the fault factors included in the fault tree are evaluated and presented, it is possible to provide a fault factor priority presenting apparatus capable of promptly specifying the fault factors.

It is a figure explaining composition of a failure factor priority presentation device in a failure tree concerning a 1st example of the present invention. FIG. 4 is a diagram illustrating a causal relationship according to the first example of the present invention. FIG. 4 is a diagram illustrating a specific causal relationship according to the first example of the present invention. FIG. 4 is a diagram illustrating an example of a field input screen according to the first embodiment of the present invention. FIG. 7 is a diagram illustrating an example of a failure tree input screen according to the first embodiment of the present invention. FIG. 5 is a diagram illustrating an example of a failure tree to be input according to the first embodiment of the present invention. FIG. 7 is a diagram illustrating an output example of the inter-event causal probability and the probability of the path from the top event to the terminal event according to the first embodiment of the present invention. It is a figure showing the example which inputted the probability of the path between the top event and the end event concerning the 1st example of the present invention. FIG. 7B is a diagram showing an example in which only factors that are equal to or greater than the threshold value input in FIG. It is a figure showing the example which displayed the causal relation concerning the 2nd example of the present invention by the number of cases.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(Equipment configuration)

FIG. 1 is a diagram illustrating a configuration of a failure factor priority presenting apparatus in a failure tree according to an embodiment of the present invention.

The failure factor priority presentation device in the failure tree has a failure factor priority presentation device 101 and a terminal device 102 in the failure tree. These can be connected via the network 103. The failure factor priority presentation device 101 in the failure tree is a general computer, and has a central control device 104 (control device), an input device 105, an output device 106, a main storage device 107, and an auxiliary storage device 108. These are interconnected by a bus. The auxiliary storage device 108 stores a causal relationship database 112 (DB) (details will be described later).

Note that the auxiliary storage device 108 is an external storage device independent of the failure factor priority presentation device 101 in the failure tree, and a configuration in which both are connected via the network 103 is also possible.

分野 The field input unit 109, the fault tree input unit 110, and the inter-event causal probability calculation unit 111 in the main storage device 107 are programs. Hereinafter, when the subject is described as “XX part”, the central control device 104 reads out each program from the auxiliary storage device 108, loads it into the main storage device 107, and then executes the function of each program (details described later). Shall be realized. The field input unit 109 receives a user's input of a field of a product / part to be evaluated for a failure tree. The fault tree input unit 110 receives an input of a fault tree by a user. The inter-event causal probability calculation unit 111 calculates an event based on the field input by the field input unit 109, the fault tree input by the fault tree input unit 110, and the causal relationship stored in the causal relationship database 112. Priority is presented by calculating the probability between the top event and the probability of the path from the top event to the terminal event. The calculation method will be described separately.

The terminal device 102 is also a general computer, and has a central control device, an input device, an output device, a main storage device, and an auxiliary storage device (not shown). These are interconnected by a bus.

The main storage device and the auxiliary storage device in the terminal device 102 may not include the programs of the main storage device 107 and the auxiliary storage device 108 in the failure factor priority presentation device 101. In this case, the failure factor may be specified by directly accessing the programs in the main storage device 107 and the auxiliary storage device 108 from the terminal device 102.
(Causal relationship)
The causal relationship database 112 stores a plurality of causal relationships. The causal relationship indicates a chain of causal effects until a product failure occurs. This causal relationship is created using information extracted from individual trouble cases that occurred in the past. The information is updated each time a failure occurs. In the first embodiment, the causal relationship is represented by a chain of a plurality of elements including parts and phenomena regarding occurrence of a product defect.

FIG. 2A is a diagram showing a causal relationship according to the first embodiment of the present invention. The causal relationship 200 is composed of three or more (plural)

elements

201a, 201b, 201c,. Each of the

elements

201a, 201b, 201c constituting the causal relationship 200 is composed of a component A202, a phenomenon A203, a component B205, a phenomenon B206, a component C208, and a phenomenon C209, respectively. The elements are connected by directed

segments

204, 207,. Of the set (two) elements connected by one directed segment, the element on the start side of the directed segment is the cause, and the element on the end point is the result. For example, element 201b is the cause as seen from element 201a. Similarly, element 201b is the result from element 201c, and element 201c is the cause from element 201b. In FIG. 2A, rosaries are expressed in a causal order, and the right element causes the left element. Note that the starting point side (cause side) of the directed line segment may be referred to as “upstream” and the end point side (result side) may be referred to as “downstream”.

In the example of FIG. 2A, the phenomenon B209 occurs in the component B205 due to the phenomenon C209 occurring in the component C208. Further, this indicates that the phenomenon A203 occurs in the component A202. Here, the component A202, the component B205, and the component C208 may be not only the component name but also a product name, a system name, and a subsystem name.

The causal relationship database 112 stores a plurality of causal relationships 200, and each of the causal relationships 200 has a causal relationship ID 200a that uniquely identifies itself. Further, a product name, a title, a representative part, and the like may be displayed together with the causal relationship ID 200a. For example, pipes as parts may be used in home appliances such as the nuclear power field and heat pump water heaters. In order to distinguish these, in addition to the causal relationship ID 200a, the product name, title, representative parts, etc. are displayed. Then, it is easy to identify.

FIG. 2B is a diagram showing a specific causal relationship according to the first example of the present invention. FIG. 2B shows an example of a causal relationship regarding “XX product cannot be activated”. The causal relationship 200 indicates the following as a whole.
A short circuit 215 occurs in the power supply circuit 214 due to the occurrence of the whisker 217 in the electronic component 216.
The output failure 213 occurs in the power supply 212 due to the short circuit 215 occurring in the power supply circuit 214.
-The start failure 211 occurs in the XX product 210 due to the output failure 213 of the power supply 212.

It is assumed that a large amount of data is stored in the causal relationship database 112 in such a manner as to the causal relationship of past defects.
(User usage procedure)
The field input section 109 displays a field input screen 301 as shown in FIG. The field input screen 301 includes a field 302 and a keyword field 303. The field 302 includes a check field 304 and a field name field 305. In the keyword column 303, keywords corresponding to the field column 302 are displayed. For example, “nuclear power or nuclear reactor or...” Is displayed in the case of the nuclear power field, and “home electrical appliance or washing machine or refrigerator” is displayed in the home electric field. Similarly, "automobile ..." is displayed in the case of the automobile field, and "PC ..." is displayed in the case of the personal computer (PC) field. This keyword is a field-specific keyword.

(4) The user selects the check box 304 for the field of the product / part to be evaluated. Then, by pressing an “OK” button 306, the selected field is determined as an evaluation target.

Next, a failure tree input screen as shown in FIG. 4 is displayed on the output device 106 by the failure tree input unit 110. FIG. 4 is a diagram showing an example of the failure tree input screen according to the first embodiment of the present invention. The user inputs the address of a file in which the fault tree to be evaluated, which is stored on the terminal device 102, is defined in the fault tree address input field 402. Here, it is assumed that the fault tree as shown in FIG. 5 is defined in the file in which the fault tree to be evaluated is defined. The failure tree may be created automatically by storing a creation program in the main storage device 107. The fault tree branches from the top event to the terminal event, and a causal relationship regarding occurrence of a product failure is displayed. FIG. 5 is a diagram showing an example of an input target fault tree according to the first embodiment of the present invention. The output device 106 displays a failure tree indicating a causal relationship of a defect related to the field input by the field input unit 109. In this failure tree, there are two factors of “piping crack” 501 in the nuclear field, “piping crack propagation” 502 and “piping corrosion” 505, and further, “factoring of piping cracking” 502 is “ There are “initial cracks in the pipe” 503 and “impact on the pipe” 504, indicating that “corrosion of the pipe” 505 is caused by “seawater inflow of the pipe” 506 and “wet state of the pipe” 507.

In evaluating the fault tree, in addition to directly inputting the address in the fault tree address input field 402, when the user presses the reference button 403, the file stored in the terminal device 102 can be displayed and selected. A screen may be displayed, and the address of the selected file may be input to the failure tree address input field 402. By pressing an “OK” button 404, the input fault tree is determined as an evaluation target.

As described above, when the field and the fault tree are input by the user, the probability of the path from the top event to the terminal event is calculated by the inter-event causal probability calculation unit 111, and the display content 601 is as shown in FIG. Is displayed on the output device 106 in a simple format. FIG. 6 is a diagram showing an output example of the inter-event causal probability and the probability of the path from the top event to the terminal event according to the first embodiment of the present invention. The method for calculating the causal probability between events will be described separately.

確率 In FIG. 6, the probabilities between events are displayed below each factor. For example, in FIG. 6, the probability that the cause of “piping breakage” 602 is “piping crack growth” 603 is displayed as 0.3, and the cause of “piping crack growth” 603 is “initial cracking of piping” 604. Is displayed as 0.2. The probability between events takes a value from 0 to 1, and the larger the value, the higher the probability of occurrence.

確率 The probability of the path between the top event and the terminal event is displayed in []. The probability of this path is displayed in the terminal event “Initial crack in piping” 604. For example, the probability from “damage to the pipe” 602 to “initial crack in the pipe” 604 indicates that 0.2 × 0.3 = 0.06. In FIG. 6, the probability of the path between the “damage of the pipe” 602 and the “wet state of the pipe” 608 is 0.63, which indicates that the probability is higher than other factors.

In the first embodiment, the probability of the terminal event “initial pipe crack” 604 is multiplied by the probability of the intermediate event “piping crack propagation” 603 before the terminal event and after the top event. Values are expressed as probabilities of the path from the top event to the terminal event. In the first embodiment, the probability 0.3 of the path between the top event “Piping breakage” 602 and the intermediate event “Piping crack growth” 603 is set to the intermediate event “Piping crack growth”. The probability 0.2 of the path between the intermediate event “Piping crack propagation” 603 and the terminal event “Initial pipe cracking” 604 is displayed in the terminal event “Initial pipe cracking”. "604. That is, the probability of the path between the top event and the intermediate event is displayed in the intermediate event. The same applies to other events.

In the end event in the first embodiment, the probability of the path between the intermediate event and the end event and the probability of the path between the top event and the end event (probability calculated by multiplication by the inter-event causal probability calculation unit 111) are described. The probability of the path between the top event and the terminal event is displayed in [] to distinguish it from the probability of the path between the intermediate event and the terminal event.

In this way, by presenting the probability of the path between the top event and the terminal event, the user can efficiently consider the factors having a high probability.
(Method of calculating causal probability between events)
A method of calculating an event-to-event causal probability will be described with an example of the probability that the cause of the “top event” is “factor A”. The probability that the cause of the “top event” is “factor A” is obtained by the following flow.
(1) The causal relationship database 112 is searched for a keyword in the field designated by the user on the field input screen 301, and a causal relationship matching the keyword condition is acquired. For example, when the field of nuclear power is specified on the field input screen 301, the causal relation database is searched with the keyword “nuclear power or nuclear reactor or...” To obtain a causal relation matching the condition.
(2) Among the causal relationships acquired in (1), a causal relationship including a “top event” is acquired, and the number of cases is counted. There may be a causal relationship in which “top event” is an end event, but here we want to find the ratio of “factor A” among the factors of “top event”. A causal relationship in which the top event is not an end event is not obtained as the number of cases.
(3) Among the causal relationships acquired in (2), a causal relationship in which the cause of the top event is “factor A” is acquired, and the number of cases is counted.
(4) The number obtained in (3) is divided by the number obtained in (2).

For example, suppose that 100 causal relationships in the nuclear field were acquired by (1). In (2), it is assumed that the causal relationship obtained in (1) includes 20 causal relationships including “top event”. Subsequently, in (3), it is assumed that two factors whose “top event” is caused by “factor A” are included. In this case, the probability that the cause of the “top event” is “factor A” is 2/20, which is 0.1.

By performing the processing of (1) to (4) above, the inter-event causal probability is obtained, and by multiplying from the top event to the terminal event, the probability of the path between the top event and the terminal event is calculated in consideration of the field. .

(4) When searching for a causal relationship, in order to absorb fluctuations in expressions such as “damaged” and “destructed”, the same word may be treated using a synonym or synonym dictionary.

According to the first embodiment, the probability of the path between the top event and the terminal event is calculated, and based on this, the priorities of the fault factors included in the fault tree are evaluated and presented. Identification can be expedited.

Furthermore, in searching for a causal relationship, a threshold value may be used. FIG. 7A is a diagram showing an example in which the probability of a path between a top event and a terminal event according to the first embodiment of the present invention is input, and FIG. 7B shows an example in which only factors exceeding the threshold value input in FIG. 7A are displayed. FIG.

As shown in FIG. 7A, the output device 106 is provided with a threshold value input screen 701. The threshold value input screen 701 is provided with an input unit 702 capable of inputting a threshold value. An arbitrary numerical value serving as a probability of a path between a top event and a terminal event is input to the input unit 702. For example, a threshold value of 0.2 is input to the input unit 702. If there is no error in the numerical values input to the input unit 702, the user presses an “OK” button 703. To cancel the operation, the user presses a “Cancel” button 704. When “OK” button 703 is pressed after inputting 0.2 in input section 702, display contents 705 are displayed on output device 106.

In FIG. 7B, for the output result of FIG. 6, only the factors having the threshold value of 0.2 or more input in FIG. 7A are displayed as a failure tree. That is, in FIG. 6, the probability of the path between the top event and the terminal event of “pipe impact” 605 is 0.24, and the probability of the path between the top event and the terminal event of “wet pipe state” 608 is 0. Since the probability is 63 and the probability of these two occurrences is 0.2 or more, a top event and a terminal event leading to “pipe impact” 605 and “pipe wet state” 608 are displayed as display contents 705. The “initial crack in the pipe” 604 and the “seawater inflow in the pipe” 607 are not displayed as the display contents 705 because the probability of the path between the top event and the terminal event is less than 0.2.

According to the first embodiment, by excluding a factor having a low probability, it is possible to more efficiently examine only a factor having a high probability.

Next, a second embodiment will be described with reference to FIG. The inter-event causal probability calculation unit 111 of the first embodiment performs the processes (1) to (4). For example, suppose that the process (1) has acquired 100 causal relationships in the nuclear field. Next, in the process (2), it is assumed that the causal relationship obtained in (1) includes 20 causal relationships including the “top event”. Subsequently, in the process (3), it is assumed that two factors whose “top event” is caused by “factor A” are included. In the process (4), the number obtained in the process (3) is divided by the number obtained in the process (2). In this case, the probability that the cause of the “top event” is “factor A” is 2/20, which is 0.1.

In the second embodiment, the probability is not calculated, but is evaluated based on the number of causal relationships in which the cause of the “top event” is “factor A”. That is, in the above example, without dividing by the number of causal relationships including the “top event”, the number of cases of the causal relationship in which the cause of the “top event” is “factor A” is “factor A”. ”. Similarly, the priority of each factor from the top event to the terminal event is calculated. An output example is shown in FIG. FIG. 8 is a diagram showing an example in which the causal relationship according to the second embodiment of the present invention is displayed by the number of cases. In FIG. 8, what is displayed by probability in FIG. 6 is displayed by the number of causal relationships. The number of causal relationships is displayed below each factor.

In the first embodiment, the probability between events is multiplied, and the probability of the path between the top event and the terminal event is displayed in [] of the terminal event. In the second embodiment, the probability is calculated and displayed. Therefore, only the number of causal relationships is calculated and displayed for each factor, and [] of the terminal event is not displayed. This processing is executed by the central control device 104 (control device).

According to the second embodiment, by displaying the number of causal cases, the user can determine the priority based on the actual number of occurrences.

In the first embodiment, the threshold value of the probability of the path between the top event and the terminal event can be input, and only the factors whose probability of the path between the top event and the terminal event is equal to or larger than the threshold value are displayed. In the example, a threshold may be input for the number of causal relationships, and only the factors above the threshold may be displayed. In this case, by excluding a factor with a low probability, it is possible to more efficiently examine only a factor with a high probability.

101: failure factor priority presentation device, 102: terminal device, 103: network, 104: central control device (control device), 105: input device, 106: output device, 107: main storage device, 108: auxiliary storage device, 109: field input unit, 110: fault tree input unit, 111: inter-event causal probability calculation unit, 112: causal relation database, 200: causal relation

Claims

A causal relationship database storing a causal relationship representing a chain of a plurality of elements including parts and phenomena regarding occurrence of a product defect;
An output device that branches from the top event to the terminal event and displays a fault tree related to the occurrence of a product defect;
An inter-event causal probability calculation unit that calculates a probability of a path from the top event to the terminal event from the causal relationship stored in the causal relationship database,
A failure factor priority presentation device, wherein the probability calculated by the inter-event causal probability calculation unit is displayed on the failure tree.
In claim 1,
Comprising an intermediate event before the terminal event and after the top event,
The inter-event causal probability calculation unit, the probability of the path of the top event and the intermediate event, the probability calculated by multiplying the intermediate event and the probability of the intermediate event is displayed on the fault tree, characterized in that Failure factor priority presenting device.
In claim 2,
A failure factor priority presentation device, wherein the probability calculated by multiplication by the inter-event causal probability calculation unit is displayed in the end event.
In claim 2,
A failure factor priority presentation device, wherein a probability of a path between the top event and the intermediate event is displayed in the intermediate event.
In claim 3,
In the end event, a probability of a path between the intermediate event and the end event and a probability multiplied and calculated by the inter-event causal probability calculation unit are written together and displayed separately. Priority presentation device.
In claim 1 or 2,
A fault factor priority presentation device, comprising: a field input unit, wherein the output device displays the fault tree indicating a causal relationship of a defect related to the field input by the field input unit.
In claim 1 or 2,
A failure factor priority presentation device, wherein the output device is provided with an input unit capable of inputting a threshold value, and displays only a factor equal to or larger than the threshold value input from the input unit as the failure tree.
A causal relationship database storing a causal relationship representing a chain of a plurality of elements including parts and phenomena regarding occurrence of a product defect;
An output device that branches from the top event to the terminal event and displays a fault tree related to the occurrence of a product defect;
A control device that calculates the number of events from the top event to the terminal event from the causal relationship stored in the causal relationship database,
A failure factor priority presentation device, wherein the number of cases calculated by the control device is displayed in the failure tree.
In claim 8,
A failure factor priority presentation device, wherein the output device is provided with an input unit capable of inputting a threshold value, and only a factor equal to or larger than the threshold value input from the input unit is displayed as the failure tree.