QUALITY CONTROL,
ASSURANCE & RELIABILITY
ET / PE ZC 434
BITS Pilani Lecture 16
Pilani Campus
RECAP – LECTURE 15
ACCEPTANCE SAMPLING PLANS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 2
TODAY’S OBJECTIVE
RELIABILITY
RECAP
DISCUSSION – COMPRE EXAM
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 3
RELIABILITY
– “Reliability is a measure of the quality of the product over the
long run.”
– We expect the product will function according to certain
expectations over a stipulated period of time
– Improve reliability in the design phase itself.
– Good reliability required for entire system including all
components - “System Reliability.”
– Find out the chances of successful operation of the product for
at least a certain stipulated period of time
– Such information helps the manufacturer to select the
parameters of a warranty policy
4
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 4
RELIABILITY
“Reliability is the probability of a product performing its intended
function for a stated period of time under certain specified
conditions”
Four aspects of Reliability
Reliability is a probability related concept
Functional performance of the product has to meet certain
stipulations
Reliability implies successful operation over a certain period of time
Operating or environmental condition under which product use takes
place are specified.
Example: Reliability of a cable is given as having a probability of
successful performance of 0.90 in withstanding 1000Kg of load for 2
years under dry condition
5
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 5
LIFE CYCLE CURVE
(BATHTUB CURVE)
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 6
LIFE CYCLE CURVE
A life cycle curve is a plot between failure rate λ and
time
Also called as ‘Bathtub curve’
Consists of three phases, namely: Debugging or
infant mortality / early failure; Chance failure /
useful life time; Wear out / ageing
Debugging phase exhibits a drop in the failure rate
as initial problems identified during prototype
testing are ironed out
In useful life phase, the failure rate is constant and
here failure occurs randomly and independently.
Also, it is called as chance failure phase.
In wear out phase, an increase in failure rate is
observed, as the product approaches its end of their
useful life as parts age and wear out
7
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 7
PROBABILITY DISTRIBUTIONS
PARAMETER OF INTEREST: FAILURE RATE
WHICH DISTRIBUTION TO USE FOR:
• USEFUL LIFE
CONSTANT FAILURE RATE – EXPONENTIAL DISTRIBUTION
• EARLY FAILURE
VARYING FAILURE RATE – WEIBULL DISTRIBUTION
• WEAR OUT FAILURE
VARYING FAILURE RATE – WEIBULL DISTRIBUTION
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 8
PROBABILITY DISTRIBUTIONS
• Exponential distribution
– During the useful failure phase, the failure rate is constant
– Hence ‘exponential distribution’ can be used to describe the time to
failure of the product for this phase
– Probability density function is given by f (t ) e t t 0
is failure rate
– Mean time to failure (MTTF) for exponential distribution is given by
MTTF 1
– If the failure rate is constant, MTTF is the reciprocal of the failure rate
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 9
RELIABILITY – TERMINOLOGY
MTBF
(OPERATING TIME)
MTTR
MTTF
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 10
RELIABILITY FUNCTION
BEHAVIOUR
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 11
RELIABILITY FUNCTION
• Exponential distribution –
– Reliability function R(t) for the exponential distribution is shown in the
figure;
– F(t) = cumulative distribution function; f(t) = probability density function
– At time 0, reliability is 1 and it decreases exponentially with time
t
R (t ) 1 F (t ) 1 e t dt e t
0
– “Failure-rate function r(t) is the ratio of the time to failure probability
density function to reliability function”
f (t ) f (t ) e t t 0
r (t )
R(t ) is failure rate
– For exponential distribution, implying a constant failure rate, then
e t
r (t ) t
e
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 12
PROBLEM
An amplifier has an exponential time-to-failure distribution with
a failure rate of 8% per 1000 hours. What is the reliability of the
amplifier at 5000 hours? Find the mean time to failure.
13
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 13
PROBLEM
What is the highest failure rate for a product if it is to have a
probability of survival (i.e., successful operation) of 95% at 4000
hours? Assume that the time to failure follows an exponential
distribution.
; loge 0.95 = loge
14
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 14
AVAILABILITY
AVAILABILITY OF A SYSTEM AT TIME “t” IS THE PROBABILITY THAT THE SYSTEM WILL BE
OPERATING AT TIME “t”
MTTF = MEAN TIME TO FAIL = 1/λ (λ = FAILURE RATE)
MTTR = MEAN TIME TO REPAIR = 1/μ (μ = REPAIR RATE)
(WHEN BOTH λ AND μ FOLLOW EXPONENTIAL DISTRIBUTION)
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 15
SYSTEM RELIABILITY
System reliability
– “Reliability of the product (made up of a number of components) is
determined by the reliability of each component and also by the
configuration of the system consisting of these components”
– Product design, manufacture, maintenance influence reliability, but
design has a major role
– One common approach for increasing the reliability of the system is
through “redundancy in design”, which is usually achieved by placing
components in parallel.
– As long as one component operates, the system operates (for parallel
systems)
16
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 16
SERIES SYSTEMS
Systems with components in series
For the system to operate, each component must operate.
It is assumed that the components operate independently of each other
(Failure of one component has no influence on the failure of any other component)
If there are ‘n’ components in series, then system reliability is given by
Rs = R1 x R2 x - - - - - - Rn
Manufacturing capability and resource limitations restrict the maximum
reliability of any given component
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 17
PROBLEM
A module of a satellite monitoring system has 500 components in series.
The reliability of each component is 0.999. Find the reliability of the module. If the
number of components in series is reduced to 200, what is the reliability of the
module?
WILL RELIABILITY INCREASE OR
DECREASE IF NO. OF COMPONENTS
DECREASES TO 200?
Rs = R1 x R2 x - - - - - - Rn
18
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 18
EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 19
EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 20
SERIES SYSTEMS
Systems with components in series
– Use of the Exponential Model
•Suppose the system has ‘n’ components in series
•Each component has exponentially distributed time-to-failure with constant
failure rates given by 1 , 2 ......n
•The system reliability is given by
1t 2 t 3t n t n
Rs e e e ...........e i t
e i 1
1
MTTF n
i 1
i
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 21
SERIES SYSTEMS
• Systems with components in series
– Use of the Exponential Model
• Thus if each component that fails is replaced immediately with
another that has the same failure rate, the mean time to failure for the
system is given by
1
MTTF n
i 1
i
• When all components have same failure rate, If i constant
then 1
MTTF
n
TIME TO FAILURE OF THE SYSTEM IS EXPONENTIALLY DISTRIBUTED WITH AN
EQUIVALENT FAILURE RATE OF
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 22
EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 23
PARALLEL SYSTEMS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 24
PARALLEL SYSTEMS
Systems with components in parallel
– System reliability can be improved by placing
components in parallel as system will operate as long
as at least one of the components operates.
– The only time the system fails is when all the parallel
components fail
– All components are assumed to operate
simultaneously.
– A system having ‘n’ components in parallel, with the
reliability of the ith component denoted by Ri, i=1, 2, --
--- n.
– Also assume that the components operate randomly
and independently of each other.
25
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 25
PARALLEL SYSTEMS
• Systems with components in parallel
– The probability of failure of each component is given by
Fi = 1-Ri.
– System fails only if all the components fail and hence the
probability of system failure is n
Fs 1 R1 1 R2 1 Rn (1 Ri )
i 1
– Reliability of the system is the complement of Fs and is given
n
by
Rs 1 Fs 1 1 Ri
i 1
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 26
PARALLEL SYSTEMS
• Systems with components in parallel
– Use of Exponential model
• In the special case, where all the components have the same failure rate the
system reliability is
n
Rs 1 1 e
i 1
i t
Rs 1 1 e
t n
• Mean time to failure for a system of n components in parallel is given by
1 1 1 1
MTTF
1 2 3 n
• MTTF for n identical components with each failed component immediately
replaced by identical component is:
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 27
EXAMPLE – PARALLEL SYSTEMS
THE RELIABILITIES OF A,B AND C ARE 0.95,0.92 AND
0.90 RESPECTIVELY. FIND THE SYSTEM RELIABILITY.
SYSTEM RELIABILITY IS HIGHER THAN THAT OF
INDIVIDUAL COMPONENTS
n
Fs 1 R1 1 R2 1 Rn (1 Ri )
i 1
n
Rs 1 Fs 1 1 Ri
i 1
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 28
EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 29
COMPLEX SYSTEMS
Complex system
– A complex system is one which has components that are both in series and in parallel
– Assumption
• Components operate independently
• Time to failure of each component is assumed to be
exponentially distributed
– The above described methods are used for calculating the reliability and failure rate
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 30
COMPLEX SYSTEMS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 31
COMPLEX SYSTEM - EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 32
COMPLEX SYSTEM - EXAMPLE
HOW TO PROCEED?
RELIABILITY VALUES GIVEN INSIDE THE BOXES
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 33
COMPLEX SYSTEM - EXAMPLE
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 34
STANDBY SYSTEM
WHAT IS THE DIFFERENCE BETWEEN
A PARALLEL SYSTEM AND A STANDBY
SYSTEM?
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 35
STANDBY SYSTEM
• IN A STANDBY SYSTEM, ONE OR MORE PARALLEL SYSTEMS
WAIT TO TAKE OVER FROM A CURRENTLY OPERATING SYSTEM
• IN A PARALLEL SYSTEM, ALL SYSTEMS ARE OPERATING AT THE
SAME TIME
• WHICH HAS HIGHER RELIABILITY – PARALLEL OR STANDBY
SYSTEMS?
• SYSTEM RELIABILITY IS HIGHER FOR STANDBY SYSTEM THAN
FOR COMPARABLE SYSTEMS OPERATING IN PARALLEL
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 36
STANDBY SYSTEM
• IF TIME TO FAIL IS EXPONENTIAL WITH FAILURE RATE λ, BY
POISSON’S DISTRIBUTION, PROBABILITY OF “x” FAILURES IN
TIME “t” IS
• FOR A STANDBY SYSTEM WITH ONE BASIC COMPONENT AND
ONE STANDBY COMPONENT IN PARALLEL, RELIABILITY
HENCE, FOR BASIC COMPONENT + “n” STANDBY COMPONENTS,
SYSTEM RELIABILITY IS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 37
EXAMPLE – STANDBY SYSTEM
FOR THE GIVEN SYSTEM (BASIC + TWO STANDBY COMPONENTS),
IF FAILURE RATE IS 0.004 / HR AND 300 HRS IS PERIOD OF
OPERATION, WHAT IS THE SYS. RELIABILITY AND MEAN TIME TO
FAILURE?
SYSTEM RELIABILITY IS:
(0.30)
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 38
EXAMPLE – STANDBY SYSTEM
INSTEAD OF STANDBY CONFIGURATION, IF ALL THREE COMPONENTS OPERATE IN
PARALLEL, THEN WHAT IS THE MTTF AND SYSTEM RELIABILITY?
STANDBY SYSTEM HAS BETTER SYSTEM RELIABILITY THAN THE PARALLEL SYSTEM
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 39
EXAMPLE – STANDBY SYSTEM
A standby system has a basic unit with four standby components.
The time to failure of each component has an exponential
distribution with a failure rate of 0.008 per hour.
– A) For a 400 hour operation period, find the reliability of the standby
system.
– B) What is its mean time to failure of the above system?
– C) Suppose all five components are operating simultaneously in
parallel, what would be the system reliability?
– D) What would be the mean time to failure for the parallel system?
40
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 40
EXAMPLE – STANDBY SYSTEM
Rs e 1 t
t t 2
t 3
.......
t
n
MTTF
n 1
2! 3! n!
A) For a 400 hour operation period, find the reliability of the standby
system.
• Rs = exp[-0.008(400)] [1 + 0.008(400) + (0.008(400))2 / 2 +
(0.008(400))3 / 6 + (0.008(400))4 / 24 = 0.7806
• B) What is its mean time to failure of the above system?
• MTTF = (n+1) / = 5 / 0.008 = 625 hours
41
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 41
EXAMPLE – STANDBY SYSTEM
(CONTD)
C) If all five units were operating in parallel, system reliability
would be
– Rs = 1 – [1 – exp(-0.008(400))5] = 0.18786
D) In that case, MTTF is
– (1/0.008) (1+ 1/2 + 1/3 + 1/4 + 1/5) = 285.4167 hours
42
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 42
Reliability and Life testing plans
Are usually destructive in nature
Involves observing a sample of items until certain number of failures occur
or observing over a certain period of time to record the number of
failures, or a combination of both
Tests are done at the prototype stage, which can be expensive depending
on the unit cost of the item
Longer test time is desirable, but needs trade-off with respect to cost of
the test plan
Test is usually conducted under simulated conditions, but it should
mimic the actual operating conditions as closely as possible
Standard plans for reliability and life testing that are used
commonly are:
– Handbook H -108
– MIL- STD - 690B
– MIL- STD - 781C
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 43
TYPES OF LIFE TESTING PLANS
FAILURE TERMINATED TEST
TIME TERMINATED TEST
SEQUENTIAL RELIABILITY TEST
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 44
RELIABILITY TESTS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 45
SEQUENTIAL TESTING
No prior decision is made as to the number of
failures or the time to conduct the test
Use accumulated results of the test to decide
whether to accept the lot, reject the lot or continue
testing
Plot graph between cumulative number of failures
for chosen sample aginst the accumulated test time
of the items
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 46
SEQUENTIAL TESTING
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 47
TOPICS COVERED SO FAR (SESSION
9 ONWARDS)
INFERENTIAL STATISTICS :
SAMPLING DISTRIBUTIONS, CENTRAL LIMIT THEOREM,
POINT AND INTERVAL ESTIMATION USING :
• Z,T AND CHI-SQUARE STATISTICS
CONFIDENCE INTERVAL OF MEAN WITH VARIANCE KNOWN (Z SCORE)
CONFIDENCE INTERVAL OF MEAN WITH VARIANCE UNKNOWN (T - DISTRIBUTION)
CONFIDENCE INTERVAL OF A VARIANCE - CHI SQUARE DISTRIBUTION
CONFIDENCE INTERVAL OF RATIO OF TWO VARIANCES – F DISTRIBUTION
HYPOTHESIS TESTING OF:
MEAN WHEN VARIANCE IS KNOWN (Z STATISTIC)
MEAN WHEN VARIANCE IS UNKNOWN (T STATISTIC)
VARIANCE (CHI-SQUARE STATISTIC)
RATIO OF TWO VARIANCES (F STATISTIC)
ERRORS IN HYPOTHESIS TESTING
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 48
TOPICS COVERED SO FAR (SESSION
9 ONWARDS)
INTRODUCTION TO CONTROL CHARTS
Issues in construction of control chart are:
Selection of control limits,
Minimizing error in inferences, Operating characteristic curve (OC curve)
Sampling size, Average Run Length (ARL)
Interpretation of control charts
CONTROL CHARTS FOR VARIABLES:
X-BAR AND R CHART
X-BAR AND s CHART
X AND MR CHART (I AND MR CHART)
FOR EACH OF THE ABOVE, BOTH NON-STANDARD AND STANDARD CASES
WERE CONSIDERED
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 49
TOPICS COVERED SO FAR (SESSION
9 ONWARDS)
CONTROL CHARTS FOR ATTRIBUTES :
‘p’ chart – proportion of non-conforming items
‘np’ chart – no. of non-conforming items
c’ chart – no. of nonconformities
‘u’ chart – no. of nonconformities per unit
‘U’ chart – no. of demerits per unit (severity of nonconformities)
PROCESS CAPABILITY ANALYSIS
Cp, Cpk, Cpm, Cpmk ratios
ACCEPTANCE SAMPLING PLANS
PRODUCER’S AND CONSUMER’S RISK,TYPES OF SAMPLING PLANS,
OC CURVE, AOQ, ATI, ASN
RELIABILITY – BEHAVIOUR, DIFFERENT TYPES OF SYSTEMS
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 50
COURSE OUTCOMES
Learning Outcomes: Students will be able to
LO1 Apply Statistical foundations to real-world situations and pertinent issues of
manufacturing and service sectors.
LO2 Recognize fundamentals of quality control, quality costs, their measurement,
analysis and various tools for quality improvement.
LO3 Explain philosophies of leading quality experts as well as Quality Awards and
Standards.
LO4 The difference between a process that is in control and a process that is
capable, through post hoc analysis of a control chart.
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 51
COURSE OUTCOMES
Learning Outcomes: Students will be able to
LO5 The relationship between Internal Failure Costs, External Failure Costs,
Appraisal Costs and Prevention Costs and how they contribute to the total
cost of quality.
LO6 Design common sampling plans; determine an appropriate sample size; set
tolerances on assemblies and components and practice acceptance sampling
procedures.
LO7 Explain Reliability and the improvement methods, selection of parameters
for warranty and use of life testing plans.
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 52
COMPRE EXAM
DISCUSS
NOTE:
PLEASE BRING NECESSARY REF. TABLES (APPENDICES FROM
TEXTBOOK T1) ALONG WITH YOU FOR EXAM. THESE WILL NOT
BE PART OF THE QUESTION PAPER SINCE THIS IS AN OPEN
BOOK EXAM.
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 53
THANK YOU
18th May’25 Quality Control, Assurance & Reliability BITS Pilani 54
