Strategy Number o F Type Modes : Annuol Symposium

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Methods For Assessing R e l i a b i l i t y Growth P o t e n t i a l

L a r r y H. Crow; US Army M a t e r i e l Systems A n a l y s i s


A c t i v i t y , Aberdeen Proving Ground

Key Words: R e l i a b i l i t y Management, R e l i a b i l i t y Growth S t r a t e g i e s , Growth P o t e n t i a l


E s t i m a t i o n , Confidence Bounds

Abstract

The r e l i a b i l i t y growth p o t e n t i a l f o r a system impact o f n o t f i x i n g c e r t a i n f a i l u r e modes and quan-


design i s d e f i n e d as t h e maximum system r e l i a b i l i t y t i f i e s t h e impact o f t h e e n g i n e e r i n g f i x e s on t h e
t h a t can be achieved f o r a p a r t i c u l a r management system growth p o t e n t i a l . Based on t h e u n d e r l y i n g
s t r a t e g y . The management s t r a t e g y determines which model, t h e s t a t i s t i c a l c o n f i d e n c e l e v e l a s s o c i a t e d
f a i l u r e modes w i l l o r w i l l n o t be f i x e d and t h e w i t h t h e r e l i a b i l i t y requirement b e i n g a t t a i n a b l e
e f f e c t i v e n e s s o f t h e f i x e s t h a t a r e incorporated. (as compared t o t h e growth p o t e n t i a l ) can be c a l c u -
I n t h i s paper we develop a p r a c t i c a l methodology l a t e d f o r a system design and management s t r a t e g y .
f o r e s t i m a t i n g t h e growth p o t e n t i a l f o r a system From these r e s u l t s , a l t e r n a t e designs, such as
based on l i m i t e d development t e s t i n g data. In high-re1 p a r t s and redundancy f o r Type A modes, can
a d d i t i o n , confidence i n t e r v a l procedures a s s o c i a t e d be assessed a l o n g w i t h e v a l u a t i n g o t h e r p o s s i b l e
w i t h t h e r e a l i s m o f t h e requirement r e l a t i v e t o t h e management s t r a t e g i e s . I f t h e r e i s h i g h r i s k asso-
growth p o t e n t i a l a r e discussed. These procedures c i a t e d w i t h meeting t h e r e l i a b i l i t y o b j e c t i v e s ,
a r e u s e f u l f o r e v a l u a t i n g t h e management s t r a t e g y then t h i s may i n d i c a t e a need t o a v o i d a c o s t l y
f o r a system and assessing t h e l i k e l i h o o d o f a t t a i n - development t e s t i n g program w i t h t h e design.
i n g the r e l i a b i l i t y objectives. I n addition, the Examples i l l u s t r a t i n g these procedures a r e g i v e n
methods a r e u s e f u l f o r t r a d e - o f f s t u d i e s e s t a b l i s h i n g i n t h e paper.
a management s t r a t e g y e a r l y i n a development program.
Numerical examples a r e g i v e n i n t h e paper i l l u s t r a t i n g Background
t h e a p p l i c a t i o n o f these procedures.
The n o t a t i o n and assumptions i n t h i s paper
Introduction p a r a l l e l those g i v e n i n Reference 1.

The i n i t i a l p r o t o t y p e s f o r a complex system It i s assumed t h a t a l l Type B modes a r e i n


w i l l g e n e r a l l y have r e l i a b i l i t y problems which must s e r i e s and f a i l i n d e p e n d e n t l y a c c o r d i n g t o t h e ex-
be found and f i x e d t h r o u g h a development t e s t i n g p o n e n t i a l d i s t r i b u t i o n . We a l s o assume t h a t t h e
program. Although t h e management s t r a t e g y toward occurrence o f Type A modes f o l l o w t h e e x p o n t e n t i a l
r e l i a b i l i t y growth may n o t be c l e a r l y d e f i n e d o r d i s t r i b u t i o n w i t h f a i l u r e r a t e AA. F i x e s f o r Type
f o r m a l l y s t a t e d , i t w i l l i n f a c t e x i s t . The growth B modes found d u r i n g t e s t may be i n c o r p o r a t e d i n t o
p o t e n t i a l f o r a system design i s defined as t h e t h e system d u r i n g t e s t o r i n c o r p o r a t e d as delayed
maximum system re1 i a b i 1 i t y t h a t can be achieved f i x e s a t t h e end o f t h e t e s t phase.
f o r a p a r t i c u l a r management s t r a t e g y . Several
elements o f t h e o v e r a l l management s t r a t e g y de- L e t K denote t h e number o f Type B modes i n t h e
system and l e t Ai be t h e f a i l u r e r a t e f o r t h e i - t h
t e r m i n e t h e growth p o t e n t i a l . F i r s t , t h e manage-
ment s t r a t e g y f o r t h e system p l a c e s f a i l u r e modes Type B mode, i = 1, ..., K. Then a t t i m e 0, t h e
i n t o two c a t e g o r i e s , Type A and Type B. Type A system f a i l u r e r a t e r ( 0 ) i s
modes a r e a l l f a i l u r e modes such t h a t when seen
d u r i n g t e s t , no c o r r e c t i v e a c t i o n w i l l be taken.
T h i s accounts f o r a l l modes f o r which i t i s de-
termined t h a t i t i s n o t c o s t - e f f e c t i v e t o attempt K
t o i n c r e a s e t h e r e l i a b i l i t y by a d e s i g n change. where X B = 1 A i . ( I n p r a c t i c e , o f course, K w i l l
Type B modes a r e a l l modes such t h a t i f seen, a i=l
design change o r f i x , w i l l be attempted. Another g e n e r a l l y n o t be known b e f o r e o r a f t e r t h e t e s t i n g . )
element o f t h e management s t r a t e g y which impacts
t h e growth p o t e n t i a l i s t h e e f f e c t i v e n e s s o f t h e D u r i n g t e s t t i m e ( 0 , t ) a random number M < K
f i x e s f o r Type B modes. The maximum r e l i a b i l i t y o f d i s t i n c t Type B modes w i l l be observed. We-further
denote by d i t h e e f f e c t i v e n e s s f a c t o r ( E F ) f o r t h e
t h a t can be achieved f o r t h e system design and
management s t r a g e g y i s a t t a i n e d when a l l Type B i - t h Type B mode, i = 1, 2, ... , K. The f a c t o r d i
i s t h e p e r c e n t decrease i n X i a f t e r a c o r r e c t i v e
modes have been seen t h r o u g h t e s t i n g and a f i x ,
w i t h a c e r t a i n e f f e c t i v e n e s s , has been i n c o r p o r - a c t i o n has been made f o r t h e i - t h Type B mode.
a t e d f o r each mode.
I f c o r r e c t i v e a c t i o n s a r e t a k e n on t h e M Type B modes
I n terms o f e s t a b l i s h i n g a t t a i n a b l e and observed by t i m e t, t h e n t h e system f a i l u r e r a t e i s
c o s t - e f f e c t i v e r e l i a b i l i t y requirments t h e f o l l o w i n g reduced from r ( 0 ) t o
q u e s t i o n i s c l e a r l y i m p o r t a n t f o r r e l i a b i l i t y growth
management: Does t h e growth p o t e n t i a l e x i s t f o r a
system design and v i a b l e management s t r a t e g i e s t o
a t t a i n t h e r e l i a b i l i t y requirements?

In this paper we g i v e p r a c t i c a l procedures t o


estimate the growth p o t e n t i a l f o r a system based
on t e s t data and f i x e f f e c t i v e n e s s and e n g i n e e r i n g
experience. The approach focuses a t t e n t i o n on t h e

U.S. Governmentwork not protectedb y US. copyright.


484 1984 PROCEEDINGS Annuol RELIABILITY AND MAINTAINABILITY Symposium
The term Again, r ( t ) represents t h e system f a i l u r e r a t e
M a t t i m e t i f f i x e s f o r a l l Type B modes observed by
t i m e T a r e incorporated i n t o t h e system w i t h e f f e c -
(3) t i v e n e s s f a c t o r s d i . The procedure f o r e s t i m a t i n g
r ( t ) i s given i n Reference (1) and i s reviewed here.

i s t h e f a i l u r e r a t e f o r t h e M modes a f t e r t h e
c o r r e c t i v e actions. The term *

J~A+AB

i
(4)

Lu
i s t h e remaining f a i l u r e r a t e f o r a l l unseen Type I-
B modes. Q
(Y

W
A l l M Type B modes observed by t e s t t i m e t may- (Y
@
not, o f course, be f i x e d by t i m e t so t h e a c t u a l
f a i l u r e r a t e a t t i m e t may n o t be r ( t ) . However,
2
r ( t ) can be viewed as t h e achieved f a i l u r e r a t e a t
9
time t i f a l l f i x e s were updated and incorporated
i n t o t h e system. See F i g u r e 1.

T
TEST TIME
Figure 2. Jump In System Failure Rate Due to
- x*+ xs Delayed Fixes .

t = TEST TIME

Figure 1. Achieved Failure Rate if Observed


Type-0 Modes Recieve Fix.

Estimation o f r ( t ) T
TEST TIME
Before we consider t h e growth p o t e n t i a l , we Figure 3. Values of Failule Rate Before and
w i l l f i r s t discuss e s t i m a t i o n procedures f o r r(t). After Delayed Fixes.
Suppose t h e system i s t e s t e d f o r a t e s t phase o f
l e n g t h T and t h a t a l l f i x e s f o r Type B modes found
d u r i n g t e s t a r e incorporated as delayed f i x e s
a t t h e end o f t h e t e s t phase. This i m p l i e s t h a t
t h e system f a i l u r e r a t e i s constant a t
Let E [ . ] denote expected value.
i t i s shown t h a t
I n Reference (1)

r(0) = ?.A t A B through t h e t e s t phase and w i l l


then jump t o a lower value r ( T ) a f t e r t h e delayed
f i x e s have been implemented. See Figure 2.

L e t N , NB be t h e t o t a l number o f Type A and


Type B f a i t u r e s observed d u r i g t h e t e s t (0,T) and
It i s a l s o noted t h a t under r e a l i s t i c assumptions,
l e t N = NA + NB. There a r e M < NB d i s t i n c t Type
6 modes observed d u r i n g t h e teSt. ( A Type B mode
i s any mode t h a t receives a f i x a t t i m e T.) A f t e r
the incorporation o f the M fixes, the f a i l u r e r a t e
f o r t h e system a t t i m e T ( a f t e r t h e jump) i s given
by t h e f u n c t i o n r ( T ) . See F i g u r e 3.

OCEEDINGS Annual BILITY AND MAINTAINABILITY Symposium 485


E[r(t)] may be expressed as From Reference [l] t h e e s t i m a t e o f h ( t ) ,
given by equation ( 6 ) i s
K
E [ r ( t ) l = XA + 1 (1-di) X i + vdh(t)
i=1

where I.’d i s t h e mean e f f e c t i v e n e s s f a c t o r and where


h ( t ) i s t h e instantaneous r a t e w i t h which a new
Type B mode w i l l occur a t t i m e t. A model o f
t h e form

B -1
h(t) = XBt
and
i s m o t i v a t e d from p r a c t i c a l experiences i n Reference
Cl]. M
A =-
Under t h e model an e s t i m a t e r ( t ) f o r r ( t ) i s
developed such t h a t
T;B
Let
E[r(t) - r ( t ) l = 0. (7) - M
d = 1 di/M.
That i s , i n t h i s sense, r ( t ) i s unbiased. i=l

For A t small, h ( t ) A t i s i n t e r p r e t e d as approx- The f a i l u r e r a t e r ( t ) i s t h e n estimated, f o r t > 0,


i m a t e l y t h e p r o b a b i l i t y t h a t a new, d i s t i n c t Type B
w i l l occur i n t h e i n t e r v a l , ( t , t + A t ) . This i s by
modelled as a non-homogeneous Poisson process. M
See F i g u r e 4. r ( t ) = NA/T + 1 1-dl) Ni/T + dh(t).
i=1

I n p a r t i c u l a r , f o r t = T, t h e end o f t h e t e s t
phase,

Example

Suppose a system i s t e s t e d f o r T = 400 hours


w i t h a t o t a l o f 42 f a i l u r e s . O f t h e 42 f a i l u r e s
1 0 were Type A (due t o modes t h a t w i l l n o t be
f i x e d ) and 32 were Type B (due t o modes t h a t w i l l
receive a f i x ) . There a r e M = 1 6 distinct f i x e s ;
t h a t i s , 16 d i s t i n c t Type B modes. L i s t e d , i n
Table 1, f o r t h e 16 modes a r e t h e corresponding
c u m u l a t i v e t e s t t i m e s when t h e y occurred. The
t = TEST TIME
f i r s t c u m u l a t i v e f a i l u r e t i m e f o r each o f t h e
16 Type B modes i s t h e t i m e o f occurrence o f a
Figure 4. Intensity Function for Distinct Type-B d i s t i n c t problem mod% t h a t i s , an X i .
Modes. Table 1. F a i l u r e Times For Type B Mc
.L
B Mode F a i l u r e Times
D u r i n g t e s t i n g over t h e i n t e r v a l (0,T) system 1 15.04, 259.99
f a i l u r e s a r e observed a t t h e N c u m u l a t i v e t e s t 2 25.26, 120.89. 366.27
...
I
times yi<y2<y3< < y ~ . O f t h e NB Type B mode 47.46; 350.2
f a i l u r e s , M o f these correspond t o d i s t i n c t Type 53.96, 315.42
B modes; t h e remainder a r e repeats. L e t X i be 56.42, 72.09, 339.97
t h e Y j which corresponds t o t h e c u m u l a t i v e t i m e 99.57, 274.71
o f t h e f i , ;t occurrence o f t h e i - t h d i s t i n c t Type 100.31
B mode, i = l , 2, ..., M. That i s , X i i s t h e cum-
u l a t i v e t e s t t i m e o f f i r s t occurrence o f t h e f i r s t
111.99, 263.47, 373.03
125.48, 164.66, 303.98
d i s t i n c t Type B mode, X2 i s t h e c u m u l a t i v e t e s t t i m e 133.43, 177.38, 324.96
o f t h e f i r s t occurrence o f t h e second d i s t i n c t Type 192.66
B mode, etc. We l e t N i denote t h e t o t a l number o f 249.1 5, 324.47
observed f a i l u r e s f o r t h e i - t h d i s t i n c t Type B mode. 13 285.01
14 379.43
For each o f t h e M d i s t i n c t Type B modes observed, 15 388.97
d i i s t h e corresponding assigned e f f e c t i v e n e s s 395.25
factors. A

486 1984 PROCEEDINGS Annual RELIABILITY AND MAINTAINABILITY Symposium


~~
~ ~~
~

From these d a t a we can determine:


problem f a i l u r e modes a r e b e i n g seen d u r i n g t e s t .
t i m e o f f i r s t occurrence T h i s r a t e d r i v e s t h e o p p o r t u n i t y t o see problems
o f t h e i - t h d i s t i n c t Type B mode. and, i n t u r n , t a k e c o r r e c t i v e a c t i o n , and, o f
course, i s an i m p o r t a n t f a c t o r i n t h e o v e r a l l
Ni - t o t a l number o f observed f a i t u w s Lor r e l i a b i l i t y growth rate. The r e l i a b i l i t y growth
p o t e n t i a l i s t h e l i m i t i n g v a l u e o f r ( t ) as t
t h e i - t h d i s t i n c t Type B mode.
increases. T h i s l i m i t i s t h e maximum r e l i a b i l i t y
For each observed Type B mode the K + , N i and assigned t h a t can be a t t a i n e d w i t h t h e management s t r a t e g y .
e f f e c t i v e n e s s f a c t o r s (EF) d i a r e l i s t e d i n Table 2. The maximum r e l i a b i l i t y w i l l be a t t a i n e d when a l l
K Type B modes have been observed and f i x e d w i t h
Table 2 F i r s t Occurrence, Number o f F a i l u r e s , and EF's di. I n terms o f f a i l u r e r a t e , t h e growth
f o r t h e Observed Type B Modes p o t e n t i a l i s expressed by t h e e q u a t i o n

K
-
Mode

1 15.04
Xi
2
Ni
- di
.67
rp = A A + 1 (i-di)
i=1
X i

2 25.26 3 .72
3 47.46 2 .?? o r , i n terms o f MTBF we have
4 53.96 2 .77
5 56.42 3 .87 Mp = l/rp. (1 6)
6 99.57 2 .92
7 100.31 1 .50 C l e a r l y , t h e management s t r a t e g y can be
8 111.99 3 .85 changed t o improved Mp. For example, Type A
9 125.48 3 .89 modes can be f i x e d t o become Type B modes.
10 133.43 4 .74 High-Re1 p a r t s o r redundancy f o r Type A modes
11 1 92.66 1 .70 c o u l d improve t h e design and reduce A A , and hence
12 249.15 2 .63 i n c r e a s e Mp. It i s i m p o r t a n t , o f course, t o
13 285.01 1 .64 be a b l e t o e s t i m a t e Mp.so t h a t t h e management
14 379.43 1 .72 s t r a t e g y can be q u a l i f i e d , and a p p r o p r i a t e
15 388.97 1 .69 t r a d e - o f f s made.
16 395.25 1 .46
E s t i m a t i o n o f t h e Growth P o t e n t i a l
The assignment'of e f f e c t i v e n e s s f a c t o r s should
be based on h i s t o r i c a l experience and e n g i n e e r i n g It i s o f i n t e r e s t t o n o t e t h a t t h e f i r s t two
c o n s i d e r a t i o n s i n regard t o t h e proposed f i x e s . terms i n ( 5 ) f o r E [ r ( t ) ] i s t h e growth p o t e n t i a l
f a i l u r e rate. We w i l l a l s o show t h a t t h e f i r s t
From equations (9), (10) and (11) we have
two terms o f r ( t ) , namely
^B = .797, = .1350 and = .721. This g i v e s
M
h(400) = .0319. I f a l l 16 f i x e s are incorporated rp = NA/T t.1 ( i - d j ) Ni/T (1 7)
i n t o t h e system a f t e r t h e 400 hour t e s t , and w i t h 1=1
t h e EF's g i v e n i n t h e t a b l e , what i s t h e new f a i l -
u r e r a t e r ( T ) ? To e s t i m a t e t h i s we use e q u a t i o n i s an unbiased e s t i m a t e o f t h e growth p o t e n t i a l
(13). The i n d i v i d u a l c a l c u l a t i o n s a r e f a i l u r e r a t e rp. To see t h i s , we observe t h a t

16 M
NA/T = 10/400 = .0250, 1 (1-di) Ni/T = .020 1 (l-di) Ni/T (18)
i=1 i=1

can be w r i t t e n as
and a h(400) = .0230. This y i e l d s t h e e s t i m a t e
r ( 4 0 0 ) = .0250 t .020 t .0230 = .068, o r a K
1 (l-di) Ni/T (1 9)
mean t i m e between f a i l u r e (MTBF) o f 14.7. i=1

The curve r ( t ) g i v e n i n F i g u r e 3 d e p i c t s t h e where, i t i s r e c a l l e d , M i s t h e observed number o f


system r e l i a b i l i t y expected a t t i m e t i f f i x e s Type B modes and K i s t h e t o t a l number o f Type B
f o r a l l Type B Modes observed t o t i m e t a r e modes. This f o l l o w s from t h e f a c t t h a t i f a Type
i n c o r p o r a t e d i n t o t h e system. To e s t i m a t e t h e B mode i s n o t seen, t h e n t h e corresponding N i i s
f u n c t i o n r ( t ) f o r t > 0 we use 0. Therefore, t h e r e a r e o n l y M non-zero terms i n
(19) and, consequently, e q u a t i o n (19) equals
r ( t ) = .0250 + .020 + ,721 h ( t ) where equat on (18). The s i g n i f i c a n c e of t h i s f a c t i s
that
-.203 M
h ( t ) = (.1350)(.797) t ,t > 0. (1 4)
EC.1
K
EC .I 1 - d i ) Ni/T] = ( 1 - d i ) Ni/T]
R e l i a b i l i t y Growth P o t e n t i a l 1=1 1=1

The f a i l u r e r a t e , r ( t ) w i l l depend on t h e manage- K K


ment s t r a t e g y which determines t h e c l a s s i f i c a t i o n =iI=1 ( - d i ) E[Ni/T] = 1 (1-dj) X i
o f Type A and Type B modes and t h e e n g i n e e r i n g 1 =1
e f f o r t which determines t h e EF's. I n addition,
r ( t ) depends on h ( t ) which i s t h e r a t e i n which

1984 PBOCEEDINGS Annual RELIABILITY A N D MAINTAINABILITY Symposium 487


~
~ ~~
~

~- ~ ~
~
Therefore, 100
rp = + 1 (1-di) X i = .0485
M K i=1
E[NA/T + 1 (1-di) Ni/T] = X A + 1 (1-di) X i . o r an MTBF o f
i=l i=1
Mp = 20.6
Thus, we may use rp t o e s t i m a t e t h e growth p o t e n t i a l
f a i l u r e r a t e . This e s t i m a t e , o f course, depends on The MTBF v a l u e 20.6 i s t h e t r u e growth p o t e n t i a l
t h e assigned EF's, Type A and Type B modes, and and t h e e s t i m a t e , based on t h e data, i s 22.2. I f the
as such, i s a u s e f u l management t o o l f o r e v a l u a t i n g i e q u i r e m e n t i s an MTBF o f 25, t h e n t h e s e r e s u l t s i n -
r e l i a b i l i t y growth s t r a t e g i e s . d i c a t e t h a t t h i s o b j e c t i v e p r o b a b l y cannot be met w i t h
t h e c u r r e n t management s t r a t e g y .
To o u t l i n e t h e procedure f o r e s t i m a t i n g t h e
g r o w t h p o t e n t i a l , suppose t h e system i s t e s t e d f o r I n p r a c t i c e , i t i s o f t e n d e s i r a b l e t o consider t h e
a p e r i o d o f t i m e T and t h a t N f a i l u r e s have been EF's ( t h a t i s , t h e e n g i n e e r i n g e f f o r t ) and t h e i r impact
observed. A c c o r d i n g t o t h e management s t r a t e g y , on t h e g r o w t h p o t e n t i a l and t h e requirement. T h i s can
NA o f t h e s e f a i l u r e s a r e Type A and NB o f t h e s e g e n e r a l l y be accomplished e f f e c t i v e l y by c o n s i d e r i n g
f a i l u r e s a r e Type B. For t h e Type B modes, t h e r e an e q u i v a l e n t average EF d, i n s t e a d o f i n d i v i d u a l
w i l l be M d i s t i n c t f i x e s . As b e f o r e , N i i s t h e di's. I n t h i s approach, we w r i t e rp as
t o t a l number o f f a i l u r e s f o r t h e i - t h Type B
mode and d i i s t h e c o r r e s p o n d i n g assigned EF. rp = A A + (1-d) (23)
From t h e s e data, t h e growth p o t e n t i a l f a i l u r e
r a t e i s e s t i m a t e d by and
M
M~ = [rP]-1
r p = NA/T +.I=1 ( 1 - d i )
1
Ni/T
For example, suppose we l e t X A = .02, Xg = .1 and
d e s i r e t h e e q u i v a l e n t average EF d such t h a t t h e MTBF
and t h e growth p o t e n t i a l MTBF i s e s t i m a t e d by e q u a l s 20.6 o r r - .0485. This would y i e l d a d
v a l u e equal t o . ? l r which i s c l o s e t o t h e t r u e popu-
iP = [;p]-l l a t i o n mean v a l u e o f .7 f o r t h e 100 d i ' S i n t h e
example. The d v a l u e would be i n d i c a t i v e o f t h e
To i l l u s t r a t e t h e a p p l i c a t i o n o f t h e e s t i m a t i o n o v e r a l l l e v e l o f e f f e c t i v e n e s s which would be
procedure f o r t h e g r o w t h p o t e n t i a l , we c o n s i d e r t h e necessary t o a t t a i n an MTBF of 20.6. If t h e r e q u i r e -
d a t a i n t h e p r e v i o u s example. I n t h i s example, T = ment were 25, t h e n under t h e same c o n d i t i o n s , a
400 and t h e f a i l u r e r a t e e s t i m a t e a t t h e end of t e s t , necessary e q u i v a l e n t average e f f e c t i v e n e s s l e v e l
a f t e r t h e i n c o r p o r a t i o n o f t h e delayed f i x e s , i s would be .8.
r ( 4 0 0 ) = .068. The c o r r e s p o n d i n g MTBF e s t i m a t e i s
14.7. To e s t i m a t e t h e maximum r e l i a b l i t y t h a t can I f a necessary e q u i v a l e n t average e f f e c t i v e n e s s
be a t t a i n e d w i t h t h i s management s t r a t e g y we use l e v e l i s t o o high, t h e n t h e o t h e r t r a d e - o f f ; t h a t
the calculations can be made a r e i n terms of Type A and Type B modes.
For an EF d, more modes w i l l have t o be f i x e d i n
NA/T = .0250 and o r d e r t o r a i s e t h e growth p o t e n t i a l . I n t h e above
example A A + X B = .12. This i s t h e i n i t i a l f a i l -
16 u r e r a t e f o r t h e system. For t h i s i n i t i a l f a i l u r e
1 (1-di) N i / T = .020. r a t e and an average d o f .71 we w i s h t o have a g r o w t h
i=1 p o t e n t i a l MTBF o f 25, o r a f a i l u r e r a t e o f .04. This
would r e q u i r e a management s t r a t e g y such t h a t X A =
From e q u a t i o n (21), t h e growth p o t e n t i a l f a i l u r e .00832 and A B = .11268. To a t t a i n t h i s , 93.9
r a t e i s e s t i m a t e d by p e r c e n t o f t h e i n i t i a l f a i l u r e r a t e must be addressed
w i t h f i x e s f o r Type B modes. T h i s compares w i t h 83.3
r p = .0250 + .020 = .045 p e r c e n t o f t h e i n i t i a l f a i l u r e r a t e when A B = .l.
and t h e growth p o t e n t i a l MTBF i s e s t i m a t e d by Confidence Bounds f o r t h e Growth P o t e n t i a l
^Mp = C.0451-l = 22.22. I f t h e g r o w t h p o t e n t i a l i s e s t i m a t e d from t h e
d a t a , t h e n i t may be d e s i r a b l e t o account f o r t h e
That i s , t h e c u r r e n t MTBF a t T = 400 f o r t h e s t a t i s t i c a l u n c e r t a i n t y o f t h i s e s t i m a t e i n comparison
system i s e s t i m a t e d t o be 14.7, and i f we c o n t i n u e t o t h e requirement. The s t a t i s t i c a l u n c e r t a i n l y can
t o do b u s i n e s s i n t h e same f a s h i o n , i t i s e s t i m a t e d be r e f l e c t e d i n terms o f c o n f i d e n c e bounds on t h e
t h a t t h e maximum MTBF t h a t can be a t t a i n e d i s 22.22. growth p o t e n t i a l .
The e s t i m a t e s r p and Mp are, o f course, s u b j e c t e d For convenience, i n t h e f o l l o w i n g discussion, l e t
t o statistical variation. I n t h i s example, t h e d a t a
A A
were generated by computer s i m u l a t i o n w i t h A A = .02, r = rp, r = rp, M = l / r , M = l / r .
Xg = .1 and K = 100. I n addition, the d j ' s f o r
each o f t h e 100 Type B modes were drawn from a Beta As n o t e d e a r l i e r , E ( r ) = r, where r i s g i v e n by
d i s t r i b u t i o n w i t h mean .7. The s i m u l a t i o n a l s o e q u a t i o n ( 2 1 ) . L e t VARC.] d e n o t e v a r i a n c e . Then,
depended on known v a l u e s o f t h e X i ' s . Based on t h e s e
parameters, t h e a c t u a l g r o w t h p o t e n t i a l f o r t h e f a i l -
ure r a t e i s

488 1984 PROCEEDINGS Annual RELIABILITY AND MAINTAINABILITY Symposium


Concl u s i o n s
VAR(;) = 1 I n t h i s paper we have e s t a b l i s h e d a framework
T i n which t h e concept o f g r o w t h p o t e n t i a l can be
c l e a r l y d e f i n e d and have p r a c t i c a l s i g n i f i c a n c e .
K With t h i s approach t h e management s t r a t e g y can be
< 1
-- [AA + 1 (l-di)ki] = r/T. q u a l i f i e d i n terms o f t h e g r o w t h p o t e n t i a l .
T i=1 Procedures were g i v e n f o r e s t i m a t i n g and p l a c i n g
c o n f i d e n c e i n t e r v a l s on t h e g r o w t h p o t e n t i a l
That i s , VAR(;) i s bounded above by r / T . Consequently, from t e s t data. Based on t h e s e methods, t r a d e -
o f f and d e c i s i o n s can be made i n r e g a r d t o t h e
t h e use o f r / T f o r t h e VAR(F) w i l l y i e l d c o n s e r v a t i v e a p p r o p r i a t e management s t r a t e g y t o f o l l o w t o
c o n f i d e n c e bounds on r. a t t a i n the r e l i a b i l i t y objectives.
If
I.. \
Biog r a p hy

D r . L a r r y H. Crow i s C h i e f o f t h e R e l i a b i l i t y
(26) Methodology O f f i c e i n t h e R e l i a b i l i t y D i v i s i o n o f
t h e US Army M a t e r i e l Systems A n a l y s i s A c t i v i t y
i s a p p r o x i m a t e l y n o r m a l l y d i s t r i b u t e d , t h e n 1 - a two- (AMSAA), Aberdeen P r o v i n g Ground, Maryland. He
s i d e d l o w e r and upper c o n f i d e n c e bounds on r a r e g i v e n o b t a i n e d h i s PhD a t F l o r i d a S t a t e U n i v e r s i t y . He
bY developed t h e AMSAA R e l i a b i l i t y Growth Model which
has been i n c o r p o r a t e d i n t o m i l i t a r y handbooks,
standards and s e r v i c e r e g u l a t i o n s on r e l i a b i l i t y .
rL = ^r + ~ 2 / 2- 44-- He c h a i r e d t h e T r i - S e r v i c e Committee t o d e v e l o p
MIL-HDBK-189, R e l i a b i l i t y Growth Management, and i s
c u r r e n t l y Chairman o f t h e US, UK, Canadian and
A u s t r a l i a n r e l i a b i l i t y group o f t h e The T e c h n i c a l
Cooporation Program (TTCP). He i s a member o f t h e
American S t a t i s t i c a l A s s o c i a t i o n and Sigma X i .
where
References
c=z/JT
1. Crow, L.H., R e l i a b i l i t y Growth P r o j e c t i o n
and Z i s t h e 1 - a / 2 p e r c e n t i l e f o r t h e normal From Delayed F i x e s , 1983 Proceedings o f t h e
d i s t r i b u t i o n . The l o w e r and upper 1 -
a con- Annual R e l i a b i l i t v and M a i n t a i n a b i l i t v
s e r v a t i v e c o n f i d e n c e bounds f o r M a r e g i v e n by Symposium, pp. 84-89.
ML = [I-,]-’, Mu = [rL1-’.
Consider a g a i n t h e d a t a used i n t h e p r e v i o u s
examples. F o r t h e s e data, T = 400 and t h e g r o w t h

p o t e n t i a l f a i l u r e r a t e e s t i m a t e i s r = .045. The

g r o w t h p o t e n t i a l MTBF e s t i m a t e i s M = 22.22. For


an 80 p e r c e n t two-sided c o n f i d e n c e bound on M we
have Z = 1.282 and C = .0641. Using equations
(27) and (28), t h e bounds on r a r e

r L = .0333 and r u = .0608.

T h e r e f o r e , t h e 80 p e r c e n t c o n f i d e n c e bounds on M
a r e ML = 16.44 and MU = 30.03.

A l o w e r bound on M i s g e n e r a l l y o f p a r t i c u l a r
interest. I n t h i s case, a 90 p e r c e n t l o w e r c o n f i d e n c e
bound on M i s 16.44. I f t h e MTBF r e q u i r e m e n t i s below
16.44, t h e n we would have h i g h s t a t i s t i c a l c o n f i d e n c e
t h a t t h e growth p o t e n t i a l e x i s t s i n t h e system
t o meet t h i s o b j e c t i v e .

489
1984 PROCEEDINGS Annual RELIABILITY AND MAINTAINABILITY Symposium

~
~~
~~ ~~
~ ~~ ~

You might also like