d_j4(μ_ci, negligible)

　

The smaller the d_ij is ,the closer theμ_ci is. Theμ_ci is just same as the jth expression if d_ij is equal to zero.

Suppose that D_ij is the smallest among the obtained distances and letη_i1η_i2,η_i3 and η_j4 represent the reciprocals of the relative distances between the identified fuzzy importance degree description μ_ci and each of defined importance degree expressions with reference to D_ij. η_ij can be defined as follows:

thenη_ij cab be normalized by:

Each θ_ij represents the extent to that eachμ_ci belongs to the jth defined importance degree expression if μ_ci completely belongs to the jth expression thenθ_ij is equal to 1 and others are to 0. The sum of values of these indices for μ_ci is equal to

confidence thatμ_ci belongs to the jth importance degree expression.

　

4. THE RESULT OF CALCULATION

　

In this section, each module importance degree is calculated based on above mentioned method. The necessary of each module can be determined based on the role of model in system, represented with fuzzy linguistic variables.

The failure degree of each module can be determined based on failure model, previously known failure rate and the affection to system of this failure, represented with fuzzy linguistic variables.

Module relation chart is drawn in Fig. 1 and based on modular action in system we can determine the influence degree and the influenced degree.

　

Fig.1 The relation chart of modules

　

Tue influence degree can be determined based on its affection items to other modules, represented with fuzzy linguistic variables. The more the items are, the bigger the influence degree is.

The influenced degree can be determined based on its affected items by other modules, represented with fuzzy linguistic variables. The more the items are, the bigger the influenced degree is.

Now we can use main engine module as an example to calculate its importance degree.

Based on its role in system the necessaryμ_n1 of main engine module can be determined as highly necessary, it can be represented by

　

[0/1 0/2 0/3 0/4 0/5 0.75/6 1/7].

Based on Fig. 1 the influence degreeμ_y1 of main engine module can be determined as average, represented by

[0/1 0.25/2 1/3 0.5/4 0/5 0/6 0/7]

　

Based on Fig.1 the influenced degree μ_b1 of main engine module can be determined as highly influenced, represented by

BACK　　　CONTENTS　　　NEXT