JPH03154536A - Network reliability analysis system - Google Patents

Abstract

PURPOSE:To display the importance of structure of each node in a network by arranging a structure importance arithmetic section calculating the structure importance relating to each node. CONSTITUTION:The system consists of a network information input section 12 receiving network information 11 comprising of number of nodes and shapes or the like of a network, a fault occurrence probability arithmetic section 13 calculating a fault occurrence probability of the network from the network information 11, a structure importance arithmetic section 15 calculating the structure importance relating to each node from the calculated fault occurrence probability and a display means 23 displaying the calculated result of the arithmetic section 15. Thus, the structural importance relating to each node of the network is displayed and the reliability of the network is improved.

Description

[Detailed Description of the Invention] "Industrial Application Field" The present invention is a network reliability analysis method used to analyze the shape of a communication network and how failures at each node affect the entire network. Regarding the system.

"Prior Art" For example, when developing a communication network, it is necessary to analyze the reliability of the network.

Network reliability analysis is performed in situations where there is no route between individual nodes (disconnected situation) when a failure occurs in any node or link that makes up the network.
It can be said that the purpose is to find out whether or not this occurs.

Such an analysis is usually treated as a problem of finding a cut set defined in graph theory.

By determining the cut set, it is possible to find combinations of node failures or link failures that cause the network to be analyzed to become disconnected, and to know the weaknesses of the network.

``Problem to be solved by the invention'' However, although conventional methods for analyzing network reliability can determine such combinations of node failures or link failures, it is difficult to determine which node is the most important. It is not possible to rank which node's failure probability will be lowered to improve the reliability of the entire network, or which node's reliability can be easily improved. This method was insufficient as a method to increase the overall performance.

Therefore, a first object of the present invention is to provide a network reliability analysis system that can display the structural importance of each node of a network.

A second object of the present invention is to provide a network reliability analysis system that can display probability importance for each node of a network.

A third object of the present invention is to provide a network reliability analysis system that can display the degree of criticality for each node of a network.

"Means for Solving the Problem" The invention according to claim 1 includes: (i) a network information input unit for inputting network information including the number of nodes, the shape of the network, etc.; (iii) a structural importance calculation unit that calculates the structural importance of each node from the calculated failure probability; and (iv) this structural importance calculation. The network reliability analysis system is provided with a display means for displaying the calculation results of the section.

Then, the structural importance calculating unit is enabled to calculate the structural importance regarding each node of the network.

Further, in the invention according to claim 2, (i) a network information input unit that inputs network information; (11) a failure probability calculation unit that calculates a failure probability of this network from the network information; (iii) A network reliability analysis system is provided with a probability importance calculating unit that calculates the probability importance for each node from the calculated probability of failure occurrence, and (iv) a display means for displaying the calculation result of the probability importance calculating unit. .

Then, the probability importance calculating unit is enabled to calculate the probability importance regarding each node of the network.

Further, in the invention according to claim 3, (i) a network information input unit for inputting network information; and (ii)
(iii) a criticality importance calculation unit that calculates the criticality importance of each node from the calculated failure probability;
(1v) The network reliability analysis system is equipped with a display means for displaying the calculation results of the criticality level calculation section.

Then, the criticality level calculation unit is enabled to calculate the criticality level for each node of the network.

"Example" Hereinafter, the present invention will be described in detail with reference to Examples.

FIG. 1 shows the configuration of a network reliability analysis system according to an embodiment of the present invention. This network reliability analysis system includes a network input section 12 for a user to input network information 11. The network input unit 12 may be configured with a keyboard as an input device, for example, or may be configured with other input means. The network information input from the network input section 12 is the FT-made lli! 28
13. The FT creation unit 13 creates an FT based on network information. The FT information 14 outputted thereby is input to a structural importance calculation section 15, a probability importance calculation section 16, and a criticality importance calculation section 17, respectively.

The structural importance calculation unit 15 calculates the structural importance regarding each node of the network. The probability importance calculating section 16 calculates the probability importance regarding each node of the network. The criticality importance calculation unit 17 calculates the criticality importance at each node of the network. These calculation results 13 to 20 are input to the result output section 23, and these results are output to the user.

Methods for outputting the results include printing them out using a printer and visually displaying them on a display.

First, F in this network reliability analysis system
The T creation section 13 will be explained.

Now, the function χi and the work state φ are defined as follows.

Here, the function χ is a function representing a basic event, that is, the state of node l. Furthermore, a network failure refers to any one point in the network excluding the failed node.
It is defined as when one-to-one communication becomes impossible. In order to calculate the probability of failure occurrence in a network, F T (Fault tree analysis), which is the basis of PTA (Fault tree analysis), is used.
) consists of this basic event, an OR or AND logic gate connecting these basic events, and the network state φ, which is the top event.

FIG. 2 shows an example of this FT.

The FT in this example is composed of a top event (φ) 31, respective basic events (χ, ) 32 to 36, and an AND gate 37.38 and an OR gate 39.40 that connect these events. Focusing on the AND gate 37 in the FT of this example, the gate output becomes "l" in a state where each node in the two basic events 32 and 33 simultaneously generates a fault. Since the OR gate 39 is placed between the AND gate 37 and the top event 310, a failure will occur in the network in the case of this simultaneous failure. 2
In a state in which one of the nodes in the two basic events 32.33 has a fault, the output of the AND gate 37 will not be 1", and the other input of the OR gate 39 will be "1".
As long as it is not 1", the network will operate normally.

Similarly, for the other two basic events, 35.3 days,
If a failure occurs in one of these nodes, the OR gate 38
The output of becomes "1". Therefore, if a failure occurs in a node corresponding to another basic event 34 at this time, a failure will occur in the network.

From the above definition, the structural importance Iφ(i
) can be expressed by the following equation (1).

Iφ(i> . . . (1) Moreover, the probability importance degree Ig (i) can be expressed by the following equation (2).

I g(i) = g(1i, q) - g(0=, (1
)...... (2) Furthermore, the criticality importance C1g (i) is as follows (3
) can be expressed by the formula.

g(q): Probability of occurrence of the top event qi: Probability of occurrence of basic event i The structural importance calculation unit 15 calculates the structural importance φ(i) according to equation (1). In addition, the probability importance calculation unit 16 calculates the probability importance 1g (i
> Calculate. Furthermore, the criticality importance calculation unit 17
Now, the criticality force and criticality importance CIg (i) are calculated according to equation (3). Such calculation results18
.about.20 will be output from the result output 923.

Example of Network Reliability Analysis Next, we will show a simple network and perform network reliability analysis on it.

FIG. 3 shows a network to be analyzed. This network consists of the first to fifth nodes “
1" to "5" and links connecting them.

The operator operates the network input section 12 shown in FIG.
, each of the 7-words “1” to “5” that make up the network.
Input network information such as the network configuration and the network shape. The FT creation unit 13 contains FTs corresponding to various basic network shapes in a ROM (read-only).
These FTs are stored in a memory (memory), and are read out by an operator's operation to create a desired FT either singly or in combination as appropriate. This creation result is displayed on a display (not shown) in the FT image creation 13, and the operator confirms the contents and, if necessary, inputs network information 11 for modification from the network input section 12 to modify the FT. conduct. Note that depending on the device, the FT may be created by analyzing network information input by the operator using graph theory, finding cut sets, and constructing a tree consisting of logic circuits.

FIG. 4 shows the result of converting the network shown in FIG. 3 into FT. This FT consists of three AND gates 61 that logically AND the basic events V+-Vs for each node "1" to "5" in a predetermined combination.
63, an OR gate 64 which takes the logical sum of these AND gates 61 to 63, and a top event 65 arranged on the output side of this OR gate 64. Here Antogame) 61 is the three basic events ￥ l V 3
The logical product of ``l' s is three 7-
This is because if all nodes 1, 3, and 5 fail, this corresponds to one of the combinations in which communication between any two points other than these becomes impossible.This is due to the previous network failure. This is derived from the definition of

AND gate 62 has three basic events y2, V<y5
This is the same reason why the AND gate 63 takes the AND of the four basic events y, .about.y. The OR gate 64 is for collecting each of these network failures into a peak event (φ) of network failure occurrence.

Calculation of structural importance Figure 5 shows the relationship between the state of each basic event Y l''-Y s and the state of the top event φ.
This is derived from the FT in FIG. Here, for each basic event y+-Ys, "1" indicates that the corresponding node is in failure, and "0" indicates that the node is operating normally.

Regarding the top event φ, when it is "l", a network failure has occurred, and when it is "0", it has not occurred.

First, the structural importance Iφ(1
).

From Figure 5, we can see that the vector (y
+ Y2 y3 y4! I's) exists in the following six ways.

(1,O,1,0,1> , (1,0,1,Ll)
, (1,L,Oll,1)<1.1,1.0.1) ,
(1, 1, 1, 1, 0), (1, 1, 1, 1, 1) Among these, the following four vectors change to φ=11 when yl=0.

(1,0,1,0,1) , (1,0,1,1,1>
, (1, 1, 1, 0, 1) (1, 1, 1, 1, O) Therefore, structural importance rφ (1) = Below, structural importance degrees φ (2) to Iφ (4) are similarly calculated. When calculated, both of these are equal to the structural importance Iφ(1)
It is April 16th.

Next, the structural importance ■φ(5) for the basic event y5
seek.

From FIG. 5, there are the following seven types of vectors where ys=1 and φ=1.

(0,1,0,1,1) , (0,1,1,1,1)
, (1,0,1,0,1)<1.0.1,1.1)
, <1.1,0,1.1) , <1.1
, 1, 0.1) (1, 1, l, 1.1) Among these, the following six vectors make φ=O when y=0.

(0,1,0,1,1) , (0,1,1,1,1>
, (1,0,1,0,1)(1,0,1,1,1>
, (1,1,0,1,1) ,(1,1,1,0,
1) Therefore, the structural importance Iφ (5) = Therefore, ■φ (5)・6/16 > I φ (1>−I φ (2)・I φ (3)
・ I φ (4)・4/16 Calculation of probability importance From Figure 5, focusing on each case where the top event is “1”, the probability of occurrence of the top event G (Q) is calculated as follows ( 4) Given by Eq.

G (q) =<1-qt)q 2(1-qt)q
tqs+ (1-Ql)Q2[13Q4Qs + Ql (1-Q2)Ql (1-Ql)qS+Q+
(1-Qz)Q3qis + QI42 (1-Ql)Q4QS +QIQ2Q3(1-Ql)QS +QIQ2Q3Q4 (1-QS> +11+QzQ3Q4qs --(4
) Also, the probability importance level 1g(i) is expressed by the above-mentioned formula (2) % Formula % In other words, the probability of each given 7-do is expressed as Ql
2) In the first term on the right-hand side of the equation, “
1", and the second term on the right side is set as "0". In this way, the probability importance 1g(i) can be calculated.

Now, the probabilities q1 to q of each node 1” to “5”.

Set each as follows.

Ql = 0.01 Q2 = 0.02 Ql = 0.03 Ql = 0.04 Qs = 0.05 At this time, the probability importance [g<1) for node 1'' is as follows.

Ig (1) = g (L, insect) - g (Obl ) = ((
1-qz)qi(1-qs)qs+(1-Qz)(hQ
aqs +Q2 (1-q3)Q4QS + Q2Q3 (1-Ql)QS + Q2Q3q4 (l QS> +QzQjq4Qs) (q2 (1-Ql) Q4QS +Qz(bQ<Qs) =0 0015 Similarly, from node "2" to node The probability importance levels 1g(2) to Ig(5) for 5"1 are as follows.

1g(2) = 0.0020 1g(3) = 0.0005 Ig(4) = 0.00105 1g(5) = 0.00109 Therefore, given the probability importance I, (i) become that way.

Ig(2) > Ig(1) > Ig(5) > I
g(4) > Ig(3) Calculation of Criticality Importance It is assumed that the probabilities q, ~q5 of each node "1" to "5" are as described above. The probability importance at this time was as follows.

1g(1) =O,OOI5 1g<2) =0.0020 Ig<3) =0.0005 Ig(4) =0.00105 Ig(5) =0.00109 Here, criticality importance C1g(i ) teeth(
3) Given by Eq. Therefore, the value of probability Q+-Qs is substituted into the denominator of the right side of equation (3), and the criticality importance level C1g(i) is determined as follows using this result.

= U, 1 do 6 8"g"'' =0.55 Xl0- Therefore, the magnitude relationship of the criticality importance C1g(i) is as follows.

Clg(5)>Clg(4)-C1g(2)>C1
g(3>=C1g(1)) The obtained magnitude relationships and the data on which they are based are displayed on the result output section 23.

``Effects of the Invention'' As described above, according to the invention as claimed in claim 1, since the structural importance calculation unit that calculates the structural importance regarding each node is arranged, the strength of each node in the network can be determined in order. Can be attached. This makes it possible to obtain useful information when changing the shape of the network.

Further, according to the invention as claimed in claim 2, since the probability importance calculating unit is arranged to calculate the probability importance for each node, reducing the failure occurrence probability of any node in the network can improve the reliability of the network. It is possible to order the devices that contribute the most, and it is possible to obtain important information when considering the order in which component devices in the network should be replaced with new devices.

Furthermore, according to the third aspect of the present invention, since a criticality importance calculation unit is arranged to calculate the criticality importance for each node, the failure occurrence probability given to the node in the network is taken into consideration. An ordering can be made as to which nodes are easiest to improve reliability in order to improve the reliability of the network.

In this way, in the present invention, the reliability of the network can be easily improved by visually recognizing analysis results such as structural importance using the display means and taking appropriate action.

[Brief explanation of the drawing]

The drawings are for explaining one embodiment of the present invention. Among them, Fig. 1 is a block diagram showing an overview of the system configuration of a network reliability analysis system, Fig. 2 is an explanatory diagram showing an example of FT, and Fig. 2 is a block diagram showing an overview of the system configuration of a network reliability analysis system. Figure 3 is a network configuration diagram showing the network to be analyzed, Figure 4 is an explanatory diagram of the network shown in Figure 3 converted to FT, and Figure 5 is the state of each basic event y+-ys and the top event. FIG. 3 is an explanatory diagram showing the relationship with the state of φ. 11...Network information, 12...Network input section, 13...
・FT creation unit, 15... Structural importance calculation unit, 16... Probability importance calculation unit, 17... Criticality importance calculation unit, 23
...Result output section, 31.65...Top event, 39.40.64...Or gate, 37
．． 38.61~63...and gate.

Claims (1)

[Claims] 1. A network information input unit that inputs network information including the number of nodes, the shape of the network, etc., and a failure probability calculation unit that calculates the probability of failure of this network from the network information. A network reliability system characterized by comprising: a structural importance calculation section that calculates the structural importance of each node from the calculated probability of failure occurrence; and a display means that displays the calculation results of the structural importance calculation section. Analysis system. 2. A network information input unit that inputs network information; a failure probability calculation unit that calculates the failure probability of this network from the network information; and a probability calculation unit that calculates the probability importance of each node from the calculated failure probability. A network reliability analysis system comprising: an importance calculation section; and a display means for displaying the calculation results of the probability importance calculation section. 3. A network information input unit that inputs network information; a failure probability calculation unit that calculates the failure probability of this network from the network information; and a failure probability calculation unit that calculates the degree of criticality for each node from the calculated failure probability. A network reliability analysis system comprising: a criticality importance calculation section; and a display means for displaying the calculation results of the criticality importance calculation section.