JP2003216238A - System and method for supporting cause analysis - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a cause analysis support system and a cause analysis supporting method capable of reducing a human load applied to cause unfolding by supporting work for investigating a cause that has brought about an event. <P>SOLUTION: A data acquiring means 4a acquires or monitors all log data related to a certain event. The acquired log data are stored in a storing part 5. If a judging means 4c judges that a change per unit time in certain log data shows a change (abrupt change, slow change, no change, or the like) that exceeds a preset threshold in the acquired log data, a selecting means 4d selects, for example the other log data. In addition, the selected data are made a graph by an output controlling means 4e and displayed on a displaying part 2. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cause analysis support system and a cause analysis support method for supporting the work of analyzing the cause of an event.

[0002]

2. Description of the Related Art In the field of engineering, there are many phenomena to be analyzed, such as vehicle speed and fuel consumption rate (fuel consumption). Generally, an event has a cause that causes the event.

For example, if deterioration of fuel consumption is observed in a performance measurement test of an automobile, it is necessary to analyze the cause. In such work, a method of manually identifying all possible causes and verifying the combination thereof by human intelligence is adopted.

By the way, an event is rarely caused by only one cause, and usually an event is caused by a plurality of causes being intertwined. The more likely the cause is, the more difficult it is to identify the cause (or combination of causes) that caused the event.

In recent years, the performance of various circuits and systems has been evaluated at the design stage by carrying out simulations. In such an application, many parameters are intricately entangled, and it is very difficult to identify the cause of the event by the conventional method.

[0006]

As described above, in recent years, it has become extremely difficult to identify the cause of an event, and it is necessary to provide some kind of support measure. The present invention has been made under the above circumstances,
It is an object of the present invention to provide a cause analysis support system and a cause analysis support method that support work for investigating the cause of an event and thereby reduce the human load required for investigating the cause.

[0007]

In order to achieve the above object, the present invention has an acquisition means for acquiring log data of a plurality of observable quantities related to a pre-defined object, and the acquisition means. Log data storage means for storing log data for each of the observable quantities, and when an event of interest occurs in any of the observable quantities, the observable quantity related to the cause of this event is selected. It is characterized by comprising a selecting means and an output means for extracting and outputting the observable amount of log data selected by the selecting means from the log data accumulating means.

In the present invention, the object may be a real existence such as a car, or an object that imitates a car by simulation using a computer.

By taking such means, the log data of the observable amount such as the fuel consumption amount and the field of view can be obtained by the obtaining device when the automobile is running, for example. The acquired log data is stored for each observable amount by the log data storage means. For example, when an event of interest such as a sudden increase in fuel consumption occurs at a certain time point, the observable quantity related to the cause of this event, such as acceleration or road surface condition, is selected by the selecting means. Then, the selected observable amount of log data is extracted from the log data accumulating unit by the output unit, subjected to processing such as graphing, and output.

With such a configuration, when an event of interest occurs, the observable quantity related to this event is automatically selected and provided to the user. Therefore, it becomes possible to assist in investigating the cause of a certain event. That is, it becomes possible to more easily carry out an analysis of the occurred event from various aspects. From this, it is possible to support the work for investigating the cause of the event, and it is possible to reduce the human load required for investigating the cause.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a functional block diagram showing the configuration of a cause analysis support system according to an embodiment of the present invention. This system is realized, for example, by loading a dedicated program on a general-purpose workstation, and its function is mainly controlled by software. This system includes an input / output unit 1 including a printer, a mouse and a keyboard (not shown), a display unit 2 such as a liquid crystal display, and an interface unit (I / F).
Part) 3, a CPU (Central Processing Unit) 4,
And a storage unit 5.

The I / F unit 3 is connected to various instruments such as a speedometer, an acceleration sensor, and a fuel gauge mounted on a vehicle via a wired or wireless line, and acts as an intermediary when data is taken into the system. Fulfill The CPU 4 executes various controls under the program stored in the storage unit 5.

In this embodiment, the CPU 4 has a data acquisition means 4a, a storage control means 4b, and a determination means 4c.
And a selection means 4d and an output control means 4e. The data acquisition unit 4a acquires log data such as vehicle speed, acceleration, fuel consumption, or the amount of visibility or road surface condition. More generally, these quantities are referred to as the observable quantities associated with a predefined object. In this embodiment,
It can also be referred to as a quantitatively measurable physical quantity related to the vehicle.

The storage control means 4b is a data acquisition means 4a.
The log data obtained in step 3 is stored and accumulated in the storage unit 5 as time passes. Log data is accumulated for each observable quantity.

The judging means 4c compares the amount of change of each log data accumulated in the storage unit 5 per unit time with a predetermined threshold value. Then, based on the result, it is determined for each observable amount whether or not a phenomenon of interest such as a rapid change has occurred. Further, when the judging means 4c judges that an event has occurred, it gives a message to the selecting means 4d to select an observable quantity related to the cause of this event.

The selecting means 4d selects an observable quantity related to the cause of this event when an event of interest occurs in any observable quantity. The output control means 4e is
The observable amount of log data selected by the selection unit 4d is read from the storage unit 5. Then, the read data is processed into a graph plotted against time, and the graph is displayed on the display unit 2
To display.

The storage unit 5 stores a fuel consumption log 5a, a speed log 5b, an acceleration log 5c, and a view log 5d in a predetermined storage area under the control of the storage control means 4b. These are the log data acquired by the data acquisition means 4a. The log data is not limited to these items, and more items may be provided. FIG. 2 is a conceptual diagram showing an example of an observable amount relating to log data that can be acquired when a vehicle is targeted. As described above, when the vehicle 6 travels, the fuel consumption, the visibility, the temperature and humidity, the road surface condition, the friction of the tire with the road surface (tire friction), the acceleration,
Observable quantities such as speed and weather may be defined. Log data relating to these observable quantities are sequentially acquired (or monitored) and accumulated in the storage unit 5.

FIG. 3 is a schematic diagram showing a graph displayed on the display 2. In this embodiment, the fuel consumption amount is taken as the observable amount in which an event has occurred. FIG. 3A is a graph showing the change over time in fuel consumption. Referring to this graph, there is a point at which a sudden change or event occurred. This is determined by the determination means 4c by a method such as comparison with a threshold value.

Then, a message is given to the selecting means 4d, and an observable amount having a causal relationship with a sudden change in the fuel consumption amount is selected by a method such as correlation processing. Here, it is assumed that the selecting unit 4d selects the amounts of speed, acceleration, field of view, tire friction, and road surface condition. Figure 3
(B) to (f) are graphs in which these amounts are plotted against time. These graphs are positioned as auxiliary graphs for investigating the cause of changes in fuel consumption.

That is, in the graph of FIG. 3 (a), there is a portion showing a sudden change. If this amount of change is larger than a preset threshold value, other data at the corresponding time is automatically graphed. This allows
Machine power assists in the work of investigating the cause of sudden changes in values.

FIG. 4 is a flow chart for explaining the operation of the system shown in FIG. In step S1 of FIG. 4, log data of each observable amount is acquired by the data acquisition means 4a, and the acquired data is acquired in step S1.
2 is stored in the storage unit 5.

In the next step S3, the determining means 4c compares the amount of change per unit time of the log data accumulated in the storage unit 5 with a predetermined threshold value. And
Based on the result, it is determined whether or not an event has occurred.

When it is determined in step S3 that an event has occurred, in the next step S4, the selecting means 4d selects an amount having a deep correlation with the event, that is, an amount having a causal relationship. Then, in step S5, a graph of the selected observable amount and the observable amount in which the event has occurred is created, and this graph is output in step S6.

As shown in FIG. 5, the above description can be applied to a more general example. FIG. 5 is a diagram showing an example of a combination of causes related to an event. For example, it is possible to consider four causes (a) to (d) for a certain event. In this case, (a),
From the combination of (b), (c), and (d), (c),
There are 10 possible combinations up to the combination of (d). Therefore, it can be inferred that one of these combinations is the true cause.

As described above, in this embodiment, all log data related to a certain event is acquired or monitored by the data acquisition means 4a. The acquired log data is stored in the storage unit 5
Accumulate in. It is determined by the determination means 4c that, among the acquired log data, a change in a certain log data per unit time indicates a change (a sharp change, a gradual change, or no change) exceeding a preset threshold. When it is determined, for example, the remaining log data is sorted by the sorting means 4d. Further, the selected data is output to the output control means 4
It was made to be a graph by e and displayed on the display unit 2.

With this arrangement, for example, when the fuel consumption of the vehicle changes abruptly, log data relating to the vehicle speed, acceleration, weather, visibility, temperature / humidity, tire friction, road surface condition, etc. are selected. , And is displayed visually. This makes it possible to analyze the cause of a certain phenomenon not only from a single point of view but also from various aspects. Moreover, since such a process is automatically performed by the CPU, it is possible to greatly reduce the time and effort of the manual process. For users, for example,
Compared to the past, the burden of thinking can be greatly reduced.

Furthermore, it is ultimately up to humans to judge what combination of causes causes the event. The system only provides the user with a combination of possible causes. Therefore, the initiative of the determination process is secured to the user side, and the user's satisfactory analysis can be performed.

As described above, according to this embodiment, the work for investigating the cause of the event is supported,
This makes it possible to reduce the human load required for investigating the cause.

The present invention is not limited to the above embodiment. For example, although the present embodiment has been described with the vehicle as the target, various other targets can be defined. Furthermore, the system of the present embodiment can be applied not only to an object that actually exists, but also to an object that imitates reality by simulation using a computer. In such an application, various data related to the simulation model are calculated by the computer. Therefore, these data can be recorded and accumulated as in the above description. Therefore, with the same processing function as above, it is possible to investigate the cause of the event that occurred in the simulation. In particular, in application to simulation, it is possible to realize an event (behavior) desired by the user by utilizing the obtained analysis result.

The graph as shown in FIG. 3 may be output not only on the display unit 2 but also on a paper medium in the form of printout, for example. Also, it is not necessary to display each graph at once. The graphs may be sequentially displayed so as to turn the pages in response to a mouse click operation.

Further, in the present embodiment, the determination means 4c is configured to determine the occurrence of an event, but the present invention is not limited to this.
The user may take the initiative in determining whether or not an event has occurred. In addition, various modifications can be made without departing from the scope of the present invention.

[0032]

As described above in detail, according to the present invention, the cause analysis support system and the cause analysis that support the work for investigating the cause of the event and can reduce the human load for investigating the cause. A support method can be provided.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a configuration of a cause analysis support system according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram showing an example of an observable amount relating to log data that can be acquired when a vehicle is targeted.

FIG. 3 is a schematic diagram showing a graph displayed on a display device 2.

FIG. 4 is a flowchart for explaining the operation of the system shown in FIG.

FIG. 5 is a diagram showing an example of a combination of causes related to an event.

[Explanation of symbols]

1 ... I / O section 2 ... Display 3 ... Interface part (I / F part) 4 ... CPU 4a ... Data acquisition means 4b ... Accumulation control means 4c ... Judgment means 4d ... sorting means 4e ... Output control means 5 ... storage unit 5a ... Fuel consumption log 5b ... speed log 5c ... Acceleration log 5d ... Sight log

Claims (8)

[Claims] 1. An acquisition unit for acquiring log data of a plurality of observable quantities related to a predefined target, and a log for accumulating the log data acquired by the acquisition unit for each of the observable quantities. Data accumulating means, judging means for judging whether or not an event of interest occurs in the plurality of observable quantities, and if the judging means judges that the event has occurred, the cause of the event is A selecting means for selecting the relevant observable quantity, and an output means for extracting the observable quantity log data selected by the selecting means from the log data accumulating means, visually processing the extracted data, and outputting the processed data. A cause analysis support system characterized by being provided. 2. The determining means compares the amount of change per unit time of the log data accumulated in the accumulating means with a predetermined threshold value for each of the plurality of observable quantities, and 2. The presence or absence of occurrence is determined.
The cause analysis support system described in. 3. The output means plots the observable log data of the event plotted against time and the observable log data related to the cause of the event with respect to time. The cause analysis support system according to claim 1, wherein the plotted graph is output. 4. The cause analysis support system according to claim 1, wherein the target is an object that imitates reality by a simulation using a computer. 5. A first step of acquiring log data of a plurality of observable quantities related to a predefined object, and a second step of accumulating the log data acquired in the first step in an accumulating means. In the plurality of observable quantities, a third step of determining whether or not an event of interest has occurred, and if it is determined that the event has occurred in this third step, it may be related to the cause of the event. It comprises a fourth step of selecting the observed amount, and a fifth step of taking out the observable amount of log data selected in the fourth step from the accumulating means, visually processing the outputted data, and outputting the processed data. A cause analysis support method characterized by the above. 6. The third step compares a change amount per unit time of the log data accumulated in the accumulating unit with a predetermined threshold value for each of the plurality of observable amounts,
The cause analysis support method according to claim 5, which is a step of determining whether or not the event has occurred. 7. The fifth step comprises a graph in which observable log data of the event is plotted against time, and observable log data related to the cause of the event is plotted against time. 6. The cause analysis support method according to claim 5, which is a step of creating and outputting a graph plotted as above. 8. The cause analysis support method according to claim 5, wherein the target is an object that imitates reality by a simulation using a computer.