Related Topics: toprightheader.gif

Crow Extended Model

Confidence Bounds Examples for the Crow Extended Model

General Examples Using the Crow Extended Model

Confidence Bounds for the Crow Extended Model

In this section, we will present the methods used in RGA to estimate the confidence bounds for the Crow Extended model when applied to developmental testing data. RGA provides two methods to estimate the confidence bounds on demonstrated MTBF and failure intensity, projected MTBF and failure intensity and growth potential MTBF and failure intensity. The two methods are the Fisher Matrix (FM) method and Crow bounds. The Fisher Matrix approach is based on the Fisher Information Matrix and is commonly employed in the reliability field. The Crow bounds were developed by Dr. Larry Crow.

This section of the on-line reference includes the following subsections:

Bounds on Demonstrated Failure Intensity for the Crow Extended Model

Fisher Matrix Bounds

If there are no BC failure modes, the demonstrated failure intensity is MATH. Thus:

MATH and: MATH

MATH (12) where MATH.

If there are BC failure modes, the demonstrated failure intensity, MATH, is actually the instantaneous failure intensity based on all of the data. λCA(T) must be positive, thus lnλCA(T)  is approximately treated as being normally distributed. MATHThe approximate confidence bounds on the instantaneous failure intensity are then estimated from: MATH (13)

where λCA(t) = λβTβ-1. MATH

The variance calculation is the same as described in the Crow-AMSAA (NHPP) chapter.

Crow Bounds

If there are no BC failure modes then:

MATH (14)

MATH (15)

where MATH.

If there are BC modes then the confidence bounds on the demonstrated failure intensity are calculated as presented in the Crow-AMSAA (NHPP) chapter.

Bounds on Demonstrated MTBF for the Crow Extended Model

Fisher Matrix Bounds

      MATH (16)
MATH
where [λD(T)]L and [λD(T)]U can be obtained from Eqn. (12).

Crow Bounds

      MATH (17)
MATH
where [λD(T)]L and [λD(T)]U can be obtained from Eqn. (14).

Bounds on Projected Failure Intensity for the Crow Extended Model

Fisher Matrix Bounds

The projected failure intensity λP(T) must be positive, thus lnλP(T) is approximately treated as being normally distributed as well: MATH MATH (18) where:

You can then get: MATH where: MATH The $Var(\hat{\beta})$ can be obtained from Fisher Matrix based on M distinct BD modes.

Crow Bounds

      MATH (19)
MATH

where MATH.

Bounds on Projected MTBF for the Crow Extended Model

Fisher Matrix Bounds

      MATH(20)
MATH

 

[λP(T)]U and [λP(T)]L can be obtained from Eqn. (18).

Crow Bounds

      MATH (21)
MATH

[λP(T)]U and [λP(T)]L can be obtained from Eqn. (19).

Bounds on Growth Potential Failure Intensity for the Crow Extended Model

Fisher Matrix Bounds

If there are no BC failure modes, the growth potential failure intensity is MATH. Then:

MATH

If there are BC failure modes, the growth potential failure intensity is MATH MATH. Therefore: MATH

The confidence bounds on the growth potential failure intensity are as follows:

      MATH (22)
MATH

where MATH.

Crow Bounds

The Crow bounds for the growth potential failure intensity are the same as the Fisher Matrix bounds.

Bounds on Growth Potential MTBF for the Crow Extended Model

Fisher Matrix Bounds

      MATH (23)
MATH

where rU and rL can be obtained from Eqn. (22).

Crow Bounds

The Crow bounds for the growth potential MTBF are the same as the Fisher Matrix bounds.