TABLE OF CONTENTS If the reliability growth trend, from past experience, is similar to the Type 1 reliability growth curve shown in Fig. 16.4, then a good model to use is the Gompertz growth model given by
| (16.1) |  |
where
0 < a ? 1, if R is in decimals,
0% < a ? 100%, if R is in percent,
0 < b < 1, 0 < c < 1, and T a > 0,
| R | = | system's or equipment's reliability at development time T a, or at launch number T a, or mission number T a, |
| a | = | upper limit that the reliability approaches asymptotically as T a ? ? (as a special case, if R ? 1 as T a ? ?, then a = 1), |
| ab | = | original level of reliability at T a = 0, |
and
| c | = | establishes the growth pattern (small values of c model rapid early reliability growth, and large values of c model slow reliability growth). |
To estimate the values of the parameters a, b, and c, do the following:
Arrange the currently available data in terms of T a and R as in Table 16.1. The T a values should be chosen at equal intervals and increasing in value by 1, such as one month, one hour, etc. For other intervals or increments, see Appendix 16A.
Calculate log R.
