Chapter 4: Exercise Questions

1. A hydraulic system is comprised of five components having constant failure rates

(days): λ1=0.001, λ2=0.005, λ3=0.0007, λ4=0.0025, and λ5=0.00001. Find the
MTTF, standard deviation and system design life if a reliability of 0.99 is desired.

2. An air conditioning compressor operates once for an average time of 20 minutes

each hour. While operating, it has experienced a failure rate of 0.01 failures per
operating hour, and while idle has experienced a dormant failure rate of 0.0002
failures per idle hour. The probability that the compressor fails on demand is 0.03.
What is the reliability of the compressor in day long operation?

3. A flashlight contains two batteries of each having an MTTF 5 operating hours

(consider constant failure rate).

• What is the probability of battery failure occurring within the first 2 operating
hours?
• If failed batteries are immediately replaced, what is the probability of more
than one failure occurring during the first 5 operating hours.

4. A microwave transmitter has exhibited a constant failure rate of 0.00034 failures

per operating hour. Which gives non-acceptable design life (0.95 reliability) of
150 hours. A second redundant transmitter is added (parallel). What is the
reliability of the system for 30 days of operation? What is design life of the new
system? What will be the hazard rate function of this redundant system?

5. A specially designed welding machine has a non-repairable motor with a constant

failure rate of 0.05 failures per year. The company has purchased two spare
motors. If the design life of the machine is 10 year, what is the probability that the
two spares will be adequate?

Reflective Questions

1. State whether the following statements are true or false.

(a). The occurrence of snow storms in St. John’s can be modeled by an

exponential distribution.
(b). The corrosion of residential hot water pipes can be modeled by constant
failure rate model.
(c). The reliability at MTTF of a component having CFR is constant.
(d). The system will have an exponential failure distribution if both
redundant components are CFR.

2. Consider a piping system having CFR, where two identical pipes are carrying
natural gas (designed to operate at 100 barg) to the refining unit. One is operating

1
 for 25 years and the other is recently replaced (with a new one). If there is a
pressure surge of 120 barg in the system, which one will fail first?

3. What is the difference between a two-parameter exponential distribution and an

exponential distribution with a burn-in period?

4. How will you determine the number of spare parts in inventory analysis if the
time between failures is exponential?

Solutions to Exercise Questions

1. CFR system with the following failure rates:

1 = 0.001 , 2 = 0.005
3 = 0.0007 , 4 = 0.0025
5 = 0.00001

Since the components follows CFR,

n
s =  i
(i). System failure rate, i =1

= 1 + 2 + 3 + 4 + 5 = 0.00921
1 1
(ii). System MTTF, MTTFs = = = 108.58days
s 0.00921
(iii). Design life for a system reliability of 0.99,

R(t ) = e−s t des =0.99

− ln(0.99)
 tdes = = 1.09days
0.00921

2. Failure on demand, with CFR

Operating time, to = 20 min .

Idle time, ti = 40 min .
Operating failure rate, o = 0.01
Idle failure rate, i = 0.0002
Failure on demand, p = 0.03

2
 ti t p
eff = i + o o +
t o + ti t o + ti t o + ti

40 20 0.03
= * 0.0002 + * 0.01 + = 0.033467 / hour
60 60 1
−  eff *t
Reliability for 24 hrs of operation, R (t ) = e
R(24) = e−0.033467*24 = 0.4479

1
3. (i). CFR System, MTTF = = 5hrs

1 1 1 1
1 = = and 2 = =
MTTF1 5 MTTF2 5
s = 1 + 2 = 2 / 5 = 0.40
Reliability for first 2 hrs of operation, R(2) = e−s t
R(2) = e−0.40*2 = 0.4493
Therefore, failure probability, F (2) = 1 − R(2)
F (2) = 1 − 0.4493 = 0.5506
(ii). If the failed batteries are immediately replaced, then the failure follows a
Poisson distribution. Therefore,

F (2) = 1 − [ P(0) + P(1)]

 1 ( t ) i   (0.2 * 5) 0 (0.2 * 5)1 

= 1 − e −t    = 1 − e −0.2*5  + 
 i =1 i!   0! 1! 
= 0.2642

4. Active redundant system with:

1 = 0.00034 and, 2 = 0.00034

R(t ) = 1 − (1 − e − t ) 2 = 2e − t − e −2t

R(t ) = 2e − 0.00034t − e − 0.00068t

Reliability over 30 days (30*24=720hrs),

3
 R(720) = 2e−0.00034*720 − e−0.00068*720 = 0.95285
1.5 1.5
MTTF = = = 4411.7hrs
 0.00034
For a single component, R1 (720) = e−0.00034*720 = 0.7828 , 22% increase in
reliability due to redundancy.

f (t ) dR(t ) 1
Hazard rate function,  (t ) = =− .
R(t ) dt R(t )

 (1 − e− t ) 0.00034(1 − e −0.00034t )
  (t ) = =
(1 − 0.5e − t ) (1 − 0.50e − 0.00034t )

5. The expected number of failures over the life (10 yrs) of machine (based on
Poisson distribution) = t = 0.05 *10 = 0.50

s
The probability of s or fewer failures over 10 yrs, Rs (t ) =  pn (t )
n =0
2
Therefore, the probability of 2 or fewer failures in 10 yrs, R2 (t ) =  pn (t )
n =0
−0.5
2
e 0.5n
R2 (t ) = 
n=0 n!
0.25
= e − 0.5 [1 + 0.50 + ] = 0.9856
2

Let Y3 be the time of third failure. Then, Y3 has a gamma distribution with k = 3
and  = 0.05 . Therefore, the expected time to obtain 3 failures is 3 / 0.05 = 60 yr .
The probability that the third failure will occur within 10 yr is obtained from,

k −1
(t ) i
FYk (t ) = 1 − e − t  , where, k = 3
i =0 i!

 0.50 0.51 0.52 

= 1 − e − 0.50  + +  = 0.0144
 0! 1! 2! 

Observe that 0.0144 = 1 − 0.9856 , the probability of 2 or fewer failures in 10 yr is

complementary to the event that the third failure occurs within 10 yr.