Outline
1 Reliability(Module - 07)
Learning Outcomes
Reliability: Basics
Special Failure Probability Distribution
Reliability of Systems
Maintainability and Availability
Reliability(Module - 07) Learning Outcomes
Basic concepts
Hazard function
Reliability of series and parallel systems
System Reliability
Maintainability
Preventive and repair maintenance
Availability
Reliability(Module - 07) Reliability: Basics
Reliability
Reliability of a component(or a system, that is, a set of components
assembled to perform a certain function) is understood as its
capability to function without break down.
The better the component performs its intended function, it is
more reliable.
The reliability is associated with dependability with successful
performance and with the absence of breakdown or failures.
In the engineering analysis point of view, it is necessary to define
the reliability quantitatively as a probability.
In the view of statistical evaluation,
Reliability is one of the important characteristics of any test.
It refers to the stability of a test measure or protocol.
It refers to the precision or accuracy of the measurement of score.
Reliability(Module - 07) Reliability: Basics
An Interesting Example on Reliability and Validity
A test designed to measure typing ability: If the test is reliable we
would expect a student who receives a high score the first time he
takes the test to receive a high score the next time he takes the
test, but score should be close. If the scores are different, it is
validity.
We can think about several other instance in real-life.
Reliability(Module - 07) Reliability: Basics
Definitions of Reliability
If a component is put into operation at some specified time, say t = 0
and if T is time until it fails or ceases to function properly, T is called
the life length or time to failure of the component. It is obvious, T ≤ 0
is a continuous random variable with some probability density function
f (t). Then, the reliability of the component at time t denoted by R(t)
is defined as
R(t) = P (T > t) = F (t) or 1 − P (T ≤ t) = 1 − F (t),
Z t
where F (t) = f (t)dt is the CDF of T.
0
Z ∞ Z t
Thus, R(t) = f (t)dt or 1 − f (t)dt.
t 0
Note: Since F (0) = 0 and F (∞) = 1 by the property of CDF,
d
R(0) = 1 and R(∞) = 0. That is, 0 ≤ R(t) ≤ 1 and also dt F (t) = f (t),
dR(t)
we get f (t) = − dt .
Reliability(Module - 07) Reliability: Basics
Hazard Function
Hazard function or instantaneous failure rate of the component
denoted by λ(t). Thus,
f (t) −R0 (t)
λ(t) = =
R(t) R(t)
Note:
Rt
R(t) = e− 0 λ(t)dt
,
Rt
f (t) = λ(t)e− 0 λ(t)dt
.
Mean and Variance
R∞
Mean Time To Failure (MTTF) = E(T ) = 0 tf (t)dt.
R∞
MTTF = 0 R(t)dt
ar(T ) = σT2 = E(T − E(T ))2 = E(T 2 ) − (E(T ))2 =
RV ∞ 2 2
0 t f (t)dt − (M T T F ) .
Reliability(Module - 07) Reliability: Basics
Conditional Reliability of a System or Component
For a wear-in period or burn-in period or after to warranty period, the
conditional reliability is defined as
R(t/T0 ) = P (T > T0 + t/T > T0 )
P (T > T0 + t)
=
P (T > T0 )
R(T0 + t)
=
R(T0 )
R T0 +t
e− 0 λ(t)dt
= R T0
e−h
0 λ(t)dt
i
R T0 +t RT
− 0 λ(t)dt− 0 0 λ(t)dt
=e
R T0 +t
− λ(t)dt
R(t/T0 ) = e T0 .
Reliability(Module - 07) Special Failure Probability Distribution
Exponential Distribution
If the time to failure T follows an exponential distribution with
parameter λ then its PDF is given by
f (t) = λe−λt , t ≥ 0.
Z ∞ h i∞
R(t) = λe−λt dt = −e−λt = e−λt .
t t
f (t) λe−λt
λ(t) = = −λt = λ.
R(t) e
which means that when the failure distribution is an exponential with
parameter λ, the failure rate at any time is a constant, equal to λ.
Conversely, when λ(t)= a constant λ, we get
Rt
f (t) = λe− 0 λdt
= λe−λt , t ≥ 0.
Due to this property, the exponential distribution is often referred to
as constant failure rate distribution in reliability contexts.
9 / 37
Reliability(Module - 07) Special Failure Probability Distribution
1
MTTF=E(T ) = λ.
V ar(T ) = σT2 = λ12 .
R(T0 +t) e−λ(T0 +t)
R(t/T0 ) = R(T0 ) = e−λT0 = e−λt
This means that the time to failure of a component is not dependent
on how long the component has been functioning. In other words, the
reliability of the component for the next 1000 hours, say is the same
regardless of whether the component is brand new or has been operat-
ing for several hours. This property is known as the memoryless prop-
erty of the constant failure rate distribution.
10 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Weibull Distribution
The PDF of the Weibull distribution was defined as
β
f (t) = αβtβ−1 e−αt , t ≥ 0.
An alternative form of Weibull’s PDF is
β−1 " #
β
β t t
f (t) = exp − , θ > 0, β > 0, t ≥ 0.
θ θ θ
By putting α = θ1β , β is called the shape parameter and θ is called the
characteristic life or scale parameter of the Weibull’s distribution. If T
follows Weibull’s distribution
β−1 " #
Z ∞ β
tβ t
R(t) = exp − dt
t θθ θ
Z ∞ β
t
= e−x dx, on putting =x
x θ
" #
β
−x t
R(t) = e = exp − .
θ
11 / 37
Reliability(Module - 07) Special Failure Probability Distribution
f (t) β t β
λ(t) = R(t) = θ θ .
MTTF = E(T ) = θΓ 1 + β1 .
h i2
2 2 2 1
Var (T) = σT = θ Γ 1 + β − Γ 1 + β .
R(t + T0 )
R(t/T0 ) =
R(T0 )
h β i
exp − t+T θ
0
= h β i
exp − Tθ0
" β #
t + T0 β
T0
R(t/T0 ) = exp − + .
θ θ
12 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Normal Distribution
If the time to failure T follows a normal distribution N (µ, σ), its PDF
is given by
−(t − µ)2
1
f (t) = √ exp , −∞ < t < ∞.
σ 2π 2σ 2
In this case,
MTTF = E(T ) = µ,
V ar(T ) = σT2 = σ 2 ,
R∞
R(t) = t f (t)dt is found out by expressing the normal integral
and using the normal tables then found out λ(t).
13 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Log-normal Distribution
If X = log(T ) follows a normal distribution N (µ, σ) then T follows
log-normal distribution whose PDF is given by
1 1
f (t) = √ exp − n o2 , t ≥ 0,
st 2π t
2s2 log tM
where s = σ is a shape parameter and tM , the median time to failure
is the location parameter given by logtM = µ. It can be proved that
2
MTTF = E(T ) = tM exp s2 .
V ar(T ) = σT2 = t2M exp(s2 ) exp(s2 ) − 1 .
14 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Example: The density function of the time to failure in years of the
gizmos(for use of widgets) manufactured by a certain company is given
200
by f (t) = (t+10)3 , t ≥ 0.
1 Derive the reliability function and determine the reliability for the
first year of operation.
2 Compute the MTTF.
3 What is the design life for a reliability 0.95?
4 Will a one-year burn-in period improve the reliability in part(1)?
If, what is the new reliability?
200
Solution: Given f (t) = (t+10)3,t ≥ 0
(1) We know that the reliability function be
−100 ∞
Z ∞
100
R(t) = f (t)dt = 2
=
t (t + 10) t (t + 10)2
100
R(1) = = 0.8264.
(1 + 10)2
15 / 37
Reliability(Module - 07) Special Failure Probability Distribution
(2)
∞ ∞
−100 ∞
Z Z
100
MTTF = R(t)dt = dt = = 10 years.
0 0 (t + 10)2 (t + 10) 0
(3) Design life is the time to failure (tD ) that corresponds to a spec-
ified reliability. Now, it is required to find tD corresponding to R =
0.95
100
= 0.95
(tD + 10)2
⇒ (tD + 10)2 = 100.2632
⇒ tD = 0.2598 years = 94.827 days ≈ 95days.
16 / 37
Reliability(Module - 07) Special Failure Probability Distribution
(4) Conditional reliability of the component with one year burn-in pe-
riod be
R(t + 1) 100 100 121
R(t/1) = = 2
÷ 2 = .
R(1) (t + 11) 11 (t + 11)2
121 10
Now R(t/1) > R(t), if (t+11)2
> (t+10)2
(t + 10)2 100
⇒ >
(t + 11)2 121
t + 10 10
⇒ >
t + 11 11
⇒ 11t > 10t which is true, t ≥ 0.
Therefore, one year burn-in period will improve the the reliability.
121
Now R(1/1) = (1+11) 2 = 0.8403 > 0.8264
17 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Example: The time to failure in operating hours of a critical solid
state power unit has the hazard rate function
t 0.5
λ(t) = 0.003 , t ≥ 0.
500
What is the reliability if the power unit must operate continuously
for 50 hours?
Determine the design life if a reliability of 0.90 is desired.
Compute the MTTF.
Given that the unit has operated for 50 hours, what is the
probability that it will survive a second 50 hours of operation?
t 0.5
Solution: Given λ(t) = 0.003h R 500 ,i t ≥ 0.
t
We know that, R(t) = exp − 0 λ(t)dt
(1)
" Z #
50
t 0.5
R(50) = exp − 0.003 dt
0 500
18 / 37
Reliability(Module - 07) Special Failure Probability Distribution
0.003 2 2/3 50
R(50) = exp − √ t
500 3 0
0.003 2 2/3
= exp − √ 50
500 3
R(50) = exp(−0.03162) = 0.9689.
(2) R(tD ) = 0.90
" Z 0.5 #
tD
t
⇒exp − 0.003 dt = 0.90
0 500
Z tD
0.003 1/2
⇒− √ t dt = −0.10536
0 500
0.003 2 3/2
⇒√ t = 0.10536
500 3 D
( √ )2/3
3 × 500 × 0.10536
∴tD = = 111.54 hours
2 × 0.003
19 / 37
Reliability(Module - 07) Special Failure Probability Distribution
(3)
Z ∞ Z ∞
0.003 2
MTTF = R(t)dt = e− √ × × t2/3 dt
500 3
Z0 ∞ 0
0.003 × 2
= e− at3/2 , where a = √
3 × 500
Z0 ∞
2
= e−x × 2/3 x−1/3 dx, on putting x = at3/2
0 3a
2 2 3
= 2/3 Γ2/3 = 2/3 Γ5/3
3a 3a 2
0.9033
= 2/3 , f rom the table of values of Gamma f unction.
a
M T T F = 45.65 hours.
20 / 37
Reliability(Module - 07) Special Failure Probability Distribution
(4)
P (T ≥ 100) R(100)
P (T ≥ 100/T ≥ 50) = =
P (T ≥ 50) R(50)
Z 100
= exp λ(t)dt
50
0.002 3/2 −0.002 3/2
= exp − √ × 100 − √ × 50
500 500
P (T ≥ 100/T ≥ 50) = exp[−0.08944 + 0.03162] = 0.9438.
Example: A cutting tool wears out with a time to failure that is nor-
mally distributed. It is known that about 34.5% of the tools fail before
9 working days and about 78.8 % fail before 12 working days.
(a) Compute the MTTF.
(b) Determine its design life for a reliability of 0.99.
(c) Determine the probability that the cutting tool will last one more
day given that it has been in use for 5 days.
21 / 37
Reliability(Module - 07) Special Failure Probability Distribution
R9
Solution: Given −∞ f (t)dt = 0.345. Let T follow a N (µ, σ)
9−µ
t−µ
Z σ
φ(z)dz = 0.345, on putting z =
−∞ σ
Z µ−9
σ
i.e., φ(z)dz = 0.155
0
µ−9
∴ = 0.4.
σ
using the normal tables. Z 12
φ(z)dz = 0.788
−∞
Z 12−µ
σ
i.e., φ(z)dz = 0.788,
−∞
Z 12−µ
σ
i.e., φ(z)dz = 0.288
0
12 − µ
∴ = 0.8.
σ
using the normal tables. 22 / 37
Reliability(Module - 07) Special Failure Probability Distribution
(ii) Let tR be the required design life for R = 0.99.
Z ∞ Z ∞
∴ f (t)dt = 0.99 or φ(z)dz = 0.99
tR −10
tR 2.5
Z 10−tR
2.5
∴ φ(z)dz = 0.49
0
10 − tR
∴ = 2.32 ⇒ ∴ tR = 4.2 days
2.5
using the normal tables.
(iii)
R∞
P (T ≥ 6) f (t)dt
P (T ≥ 6/T > 5) = = R6∞
P (T > 5) 5
f (t)dt
R∞ R 1.6
−1.6
φ(z)dz 0.5 + 0 φ(z)dz
= R∞ = R2
−2
φ(z)dz 0.5 + 0 φ(z)dz
0.94520
P (T ≥ 6/T > 5) = = 0.9672.
0.97725
23 / 37
Reliability(Module - 07) Special Failure Probability Distribution
Examples:
A relay circuit has an MTBF of 0.8 year. Assuming random
failures. (a) Calculate the probability that the circuit will survive
1 year without failure, (b) What is the probability that there will
be more than 2 failures in the first year? (c) What is the expected
number of failures per year?
The reliability of a turbine blade is given by
2
R(t) = 1 − tt0 , 0 ≤ t ≤ t0 , where t0 is the maximum life of the
blade. (a) Show that the blades are experiencing wear out, (b)
Compute MTTF as a function of the maximum life, (c) If the
maximum life is 2000 operating hours, what is the design life for a
reliability of 0.90?
√
Given that R(t) = e− 0.001t , t ≥ 0. (a) Compute the reliability for
a 50 hours mission, (b) Show that the hazard rate is decreasing,
(c) Given a 10-hour wear-in period, compute the reliability for a
50 hour mission, (d) What is the design life for a reliability of
0.95, given a 10 hour wear-in period?
24 / 37
Reliability(Module - 07) Reliability of Systems
Reliability of Systems
A system is generally understood as a set of components
assembled to perform a certain function.
In the previous section we have considered a number of important
failure models (distributions) for the individual components.
To evaluate the reliability of a complex system, we may apply a
particular-failure law to the entire system. But it will be proper if
we determine an appropriate reliability model for each component
and then compute the reliability of the system by applying the
relevant rules of probability according to the configuration of the
components within the system.
In this section, we shall discuss the system reliability in respect of
a few simple but relatively important cases.
Serial configuration
Parallel configuration
25 / 37
Reliability(Module - 07) Reliability of Systems
Serial Configuration
Series or non-redundant configuration is one in which the components
of the system are connected in series (or serially) as shown in the fol-
lowing reliability block diagram. Each block represents a component.
In series configuration, all components must function for the system to
function. In other words the failure of any component causes system
failure.
Let R1 (t), R2 (t) and Rs (t) be the reliabilities of the components C1
and C2 and the system(assuming that there are only 2 components in
series). Then
R1 = P (C1 ) = probability that Cl functions and
R2 = P (C2 ) = probability that C2 functions. Now
Rs = probability that both C1 and C2 function.
26 / 37
Reliability(Module - 07) Reliability of Systems
Rs = P (C1 ∩ C2 ) = P (C1 ). P (C2 ), assuming that C1 and C2 function
independently Rs = R1 × R2 .
This result may be extended. If C1 , C2 , · · · , Cn be a set of n indepen-
dent components in series with reliabilities R1 (t), R2 (t), · · · , Rn (t)
Rs (t) = R1 (t) × R2 (t) × · · · × Rn (t)
Rs (t) ≤ min{R1 (t), R2 (t), · · · , Rn (t)}[∵ 0 < Ri (t) < 1]
That is, the system reliability will not be greater than the smallest of
the component reliabilities.
Note:
If each component has a constant failure rate λi , then
Rs (t) = e−λ1 t .e−λ2 t · · · e−λn t = e−(λ1 +λ2 +···+λn )t = e−λs t .
If the components follow the Weibull failure law the parameters βi
and θi , then
" # " # " n #
β1 βn X t βi
t t
Rs (t) = exp − × · · · × exp − = exp −
θ1 θn i=1
θi
This means, the system does not follow Weibull failure law, even
though every component follows a Weibull failure distribution.
27 / 37
Reliability(Module - 07) Reliability of Systems
Parallel Configuration
Parallel or redundant configuration is one in which the components of
the system are connected in parallel as shown in the following reliabil-
ity block diagram
In parallel configuration, all components must fail for the system to
fail. This means that if one or more components function, the system
continues to function.
28 / 37
Reliability(Module - 07) Reliability of Systems
Taking n = 2 and denoting the system reliability by Rp , we have
Rp = P (C1 or C2 or bothf unction) = P (C1 ∪ C2 )
= P (C1 ) + P (C2 ) − P (C1 ∩ C2 )
= P (C1 ) + P (C2 ) − P (C1 )P (C2 ), C1 , C2 are independent
Rp = R1 + R2 − R1 R2 = 1 − (1 − R1 )(1 − R2 ).
Extending to n components, we have
Rp = 1 − (1 − R1 )(1 − R2 ) · · · (1 − Rn ) ≥ M ax{R1 .R2 , · · · , Rn }.
Note: For a two component system in parallel having constant failure
rate Ri = 1 − e−λi t , i = 1, 2.
Hence,
Rp (t) = 1 − (1 + e−λ1 t )(1 − e−λ2 t ) = e−λ1 t + e−λ2 t − e−(λ1 +λ2 )t
and Z ∞
1 1 1
MTTF = Rp (t)dt = + − .
0 λ1 λ2 λ1 + λ2
29 / 37
Reliability(Module - 07) Reliability of Systems
Parallel-series Configuration
A system, in which m subsystems are connected in series where each
sub-system has n components connected in parallel as shown in the
following reliability block diagram is said to be parallel-series configu-
ration or low-level redundancy.
If R is the reliability of the individual component, the reliability of
each of the sub-system be
1 − (1 − R)n , Here, n = 2.
Since m sub-systems are connected in series(m = 3), the system relia-
bility for the low-level redundancy is given by
RLow = [1 − (1 − R)n ]m .
30 / 37
Reliability(Module - 07) Reliability of Systems
Series-parallel Configuration
A system, in which m subsystems are connected in parallel where each
sub-system has n components connected in series as shown in the fol-
lowing reliability block diagram is said to be parallel-series configura-
tion or high-level redundancy.
If R is the reliability of each component, the reliability of each of the
sub-system be Rn , here, n = 2.
Since m sub-systems are connected in parallel(m = 3), the system
reliability for the high-level redundancy is given by
RHigh = 1 − (1 − Rn )m .
Note: When m = n = 2, RLow ≥ RHigh , since
RLow − RHigh = 2R2 (1 − R)2 ≥ 0.
31 / 37
Reliability(Module - 07) Reliability of Systems
Example: An electronic circuit consists of 5 silicon transistors, 3 sil-
icon diodes, 10 composition resistors and 2 ceramic capacitors con-
nected in series configuration. The hourly failure rate of each compo-
nent is given below
1 Silicon transistor : λ = 4 × 10−05 .
t
2 Silicon diode: λ = 3 × 10−05 .
d
3 Composition resistor: λ = 2 × 10−04 .
r
4 Ceramic capacitor : λ = 2 × 10−04 .
c
Calculate the reliability of the circuit for 10 hours, when the compo-
nents follow exponential distribution. [Since the component are con-
nected in series]
Solution:
Rs (t) = R1 (t).R2 (t).R3 (t).R4 (t) = e−λ1 (t) .e−λ2 (t) .e−λ3 (t) .e−λ4 (t)
= e−(5λt +3λd +10λr +2λc )t
−05 +9×10−05 +20×10−04 +4×10−04 )×10
Rs (10) = e−(20×10
−04
Rs (10) = e−(20+9+200+40)×10 = e−0.0269 = 0.9735.
32 / 37
Reliability(Module - 07) Reliability of Systems
Example: A system consists of two sub systems in parallel. The re-
t 2
liability of each subsystem is given by R(t) = e−( θ ) . Determine the
system MTTF.
Solution: If R(t) is the reliability function of the engine(system), then
t 2 2
h i
Rp (t) = 1 − 1 − e−( θ )
2 2
− t2 − 2t2
= 1 − 1 − 2e θ + e θ
t2 2t2
−
Rp (t) = 2e θ2 −e θ2 .
System MTTF be
Z ∞ Z ∞ 2
Z ∞ 2
− t2 −2 t 2
Rp (t)dt = 2 e θ dt − e θ dt = 1.15θ.
0 0 0
33 / 37
Reliability(Module - 07) Reliability of Systems
Example: A Jet engine consists of 5 modules(connected in series)
each of which was found to have a weibull failure distribution with a
shape parameter 1.5. Their characteristic lives are(in operating cycles)
3600, 7200, 5850, 4780 and 9300. Find the reliability function of the
engine and the MTTF.
Solution: If R(t) is the reliability function of the engine(system), then
Rs (t) = R1 (t).R2 (t).R3 (t).R4 (t).R5 (t)
1.5 1.5 1.5
− θt − t
− θt
=e 1 .e θ2
···e 5
−1.5
+θ2−1.5 +···+θ5−1.5 )t1.5
= e−(θ1
1.5 t 1.5
Rs (t) = e−0.000012642t = e−( 1842.7 ) , where θ = 1842.7, β = 1.5.
MTTF of the eigne be
1 1
θΓ 1 + = 1842.7 × Γ 1 +
β 1.5
1
θΓ 1 + = 1664.5 cycles.
β
34 / 37
Reliability(Module - 07) Maintainability and Availability
No equipment(system) can be perfectly refinable in spite of the utmost
care and best effort on the part of the designer and manufacturer. In
fact, very few systems are designed to operate without maintenance of
any kind. For a large number of systems, maintenance is must, as it is
one of the effective ways o
35 / 37
