QUALITY ENGINEERING
Lecture 9
“Reliability”
Presented by:
Dr. Mahendran Samykano
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
OUTLINE
FUNDAMENTAL ASPECTS
ADDITIONAL STATISTICAL ASPECTS
LIFE AND RELIABILITY TESTING PLANS
TEST DESIGN
AVAILABILITY AND MAINTAINABILITY
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc2
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LEARNING OBJECTIVES
WHEN YOU HAVE COMPLETED THIS CHAPTER YOU SHOULD BE
ABLE TO:
KNOW THE DEFINITION OF RELIABILITY AND THE FACTORS
ASSOCIATED WITH IT.
KNOW THE VARIOUS TECHNIQUES TO OBTAIN RELIABILITY.
UNDERSTAND THE PROBABILITY DISTRIBUTIONS, FAILURE CURVES,
AND RELIABILITY CURVES AS A FACTOR OF TIME.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc3
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LEARNING OBJECTIVES
WHEN YOU HAVE COMPLETED THIS CHAPTER YOU SHOULD
BE ABLE TO:
CALCULATE THE FAILURE RATE UNDER DIFFERENT CONDITIONS.
CONSTRUCT THE LIFE HISTORY CURVE AND DESCRIBE ITS THREE
PHASES.
CALCULATE THE NORMAL, EXPONENTIAL, AND WEIBULL FAILURE
RATE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc4
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LEARNING OBJECTIVES
WHEN YOU HAVE COMPLETED THIS CHAPTER YOU SHOULD BE
ABLE TO:
CONSTRUCT THE OC CURVE
DETERMINE LIFE AND RELIABILITY TEST CURVES
CALCULATE THE NORMAL, EXPONENTIAL, AND WEIBULL FAILURE RATE
UNDERSTAND THE DIFFERENT TYPES OF TEST DESIGN
UNDERSTAND THE CONCEPTS OF AVAILABILITY AND MAINTAINABILITY
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc5
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
RELIABILITY
GENERALLY DEFINED AS THE ABILITY OF A PRODUCT TO PERFORM
AS EXPECTED OVER TIME.
FORMALLY DEFINED AS THE PROBABILITY THAT A PRODUCT, PIECE
OF EQUIPMENT, OR SYSTEM WILL PERFORM ITS INTENDED FUNCTION
FOR A PRESCRIBED LIFE UNDER STATED ENVIRONMENTAL
CONDITIONS
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc6
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
RELIABILITY
MEANS QUALITY OVER THE LONG RUN.
A PRODUCT THAT “WORKS” FOR A LONG PERIOD OF
TIME IS A RELIABLE ONE.
SINCE ALL UNITS OF A PRODUCT WILL FAIL AT DIFFERENT
TIMES, RELIABILITY IS A PROBABILITY.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc7
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
RELIABILITY
THERE ARE FOUR FACTORS ASSOCIATED WITH RELIABILITY:
1. NUMERICAL VALUE.
THE NUMERICAL VALUE IS THE PROBABILITY THAT THE PRODUCT
WILL FUNCTION SATISFACTORILY DURING A PARTICULAR TIME.
2. INTENDED FUNCTION.
PRODUCT ARE DESIGNED FOR PARTICULAR APPLICATIONS AND
ARE EXPECTED TO BE ABLE TO PERFORM THOSE APPLICATIONS.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc8
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
RELIABILITY
THERE ARE FOUR FACTORS ASSOCIATED WITH RELIABILITY:
3. LIFE.
HOW LONG THE PRODUCT IS EXPECTED TO LAST. PRODUCT LIFE IS
SPECIFIED AS A FUNCTION OF USAGE, TIME, OR BOTH.
4. ENVIRONMENTAL CONDITIONS
INDOORS.
OUTDOORS.
STORAGE.
TRANSPORTATION.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc9
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
ACHIEVING RELIABILITY
EMPHASIS:
1. THE CONSUMER PROTECTION ACT OF 1972.
2. PRODUCTS ARE MORE COMPLICATED.
3. AUTOMATION.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc10
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
SYSTEM RELIABILITY
AS PRODUCTS BECOME MORE COMPLEX (HAVE MORE COMPONENTS),
THE CHANCE THAT THEY WILL NOT FUNCTION INCREASES.
THE METHOD OF ARRANGING THE COMPONENTS AFFECTS THE
RELIABILITY OF THE ENTIRE SYSTEM.
COMPONENTS CAN BE ARRANGED IN SERIES, PARALLEL, OR A
COMBINATION.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc11
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
SERIES SYSTEM
For a series systems, the reliability is the product of the individual
components.
1 2 n
RS = R1 R2 ... Rn
As components are added to the series, the system reliability
decreases.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc12
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
PARALLEL SYSTEM
1
2
n Rs = 1 - (1 - R1) (1 - R2)... (1 - Rn)
When a component does not function, the product continues to function,
using another component, until all parallel components do not function.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc13
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
SERIES-PARALLEL SYSTEM
C
RA RB RC RD
A B D
C
RC
CONVERT TO EQUIVALENT SERIES SYSTEM
RA RB RD
A B C’ D
RC’ = 1 – (1-RC)(1-RC)
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc14
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
DESIGN
THE MOST IMPORTANT ASPECT OF RELIABILITY IS THE DESIGN.
IT SHOULD BE AS SIMPLE AS POSSIBLE.
THE FEWER THE NUMBER OF COMPONENTS, THE GREATER THE
RELIABILITY.
ANOTHER WAY OF ACHIEVING RELIABILITY IS TO HAVE A BACKUP
OR REDUNDANT COMPONENT (PARALLEL COMPONENT).
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc15
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
DESIGN
RELIABILITY CAN BE ACHIEVED BY OVERDESIGN.
THE USE OF LARGE FACTORS OF SAFETY CAN INCREASE THE
RELIABILITY OF A PRODUCT.
WHEN AN UNRELIABLE PRODUCT CAN LEAD TO A FATALITY OR
SUBSTANTIAL FINANCIAL LOSS, A FAIL-SAFE TYPE OF DEVICE SHOULD
BE USED.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc16
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
PRODUCTION
THE SECOND MOST IMPORTANT ASPECT OF RELIABILITY IS THE
PRODUCTION PROCESS.
EMPHASIS SHOULD BE PLACED ON THOSE COMPONENTS WHICH
ARE LEAST RELIABLE.
PRODUCTION PERSONNEL.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc17
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
TRANSPORTATION
THE THIRD MOST IMPORTANT ASPECT OF RELIABILITY IS THE
TRANSPORTATION.
PACKAGING
SHIPMENT
PERFORMANCE OF THE PRODUCT BY THE CUSTOMER IS THE FINAL
EVALUATION.
GOOD PACKAGING TECHNIQUES AND SHIPMENT EVALUATION
ARE ESSENTIAL.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc18
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
MAINTENANCE
ELIMINATE NEED
WARNINGS, SUCH AS AUDIBLE OR VISUAL SIGNALS
SIMPLE AND EASY TO PERFORM
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc19
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
ADDITIONAL STATISTICAL ASPECTS
DISTRIBUTIONS APPLICABLE TO RELIABILITY:
EXPONENTIAL DISTRIBUTION.
NORMAL DISTRIBUTION.
WEIBULL DISTRIBUTION.
RELIABILITY CURVES:
THE CURVES AS A FUNCTION OF TIME.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc20
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
ADDITIONAL STATISTICAL ASPECTS
RELIABILITY CURVES:
THE RELIABILITY CURVES FOR THE EXPONENTIAL, NORMAL
AND WEIBULL DISTRIBUTIONS AS A FUNCTION OF TIME ARE
GIVEN IN FIGURE 11-2(B) .
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc21
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc22
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
ADDITIONAL STATISTICAL ASPECTS
FAILURE-RATE CURVE:
IT IS IMPORTANT IN DESCRIBING THE LIFE-HISTORY CURVE OF A
PRODUCT.
SEE FIGURE 11-2C.
number of test failures r
λest = =
sum of test times t +(n - r)T
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc23
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
EXPONENTIAL NORMAL WEIBULL
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc24
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE HISTORY CURVE
The curve, sometimes referred to as the “bathtub”
curve, is a comparison of failure rate with time.
It has three distinct phases:
The debugging phase.
The chance failure phase.
The wear-out phase.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc25
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE HISTORY CURVE
Wear Out
Chance Failure Phase
“Infant Debugging
Phase
Phase
mortality
period”
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc26
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE HISTORY CURVE
1. The debugging phase:
It is characterized by marginal and short-life parts that cause a
rapid decrease in the failure rate.
It may be part of the testing activity prior to shipment for some
products.
The Weibull distribution ß<1 is used to describe the occurrence of
failures.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc27
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE HISTORY CURVE
2. The chance failure phase:
Failures occur in a random manner due to the constant failure rate. The
Exponential and the Weibull distributions β= 1 are best suited to
describe this phase.
3. The wear-out phase:
Is depicted by a sharp raise in failure rates. The Normal distribution
and the Weibull distribution ß >1 are used to describe this phase.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc28
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
NORMAL FAILURE ANALYSIS
THE WEIBULL DISTRIBUTION IS USUALLY USES.
THE NORMAL DISTRIBUTION.
t
R(t) = 1.0 - f(t)dt
0
R(t) = 1.0 - P(t)
R(T): RELIABILITY AT TIME T
P(T): PROBABILITY OF FAILURE OR AREA OF THE NORMAL
CURVE TO THE LEFT OF TIME T. TABLE A.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc29
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
EXPONENTIAL FAILURE ANALYSIS
EXPONENTIAL DISTRIBUTION:
RT = E –T/Ө
WHERE:
T: TIME OR CYCLES.
Ө: MEAN LIFE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc30
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
WEIBULL FAILURE ANALYSIS
CAN BE USED FOR THE DEBUGGING PHASE (ß<1) AND THE
CHANCE FAILURE PHASE (ß=1).
BY SETTING = 1, THE WEIBULL EQUALS THE EXPONENTIAL.
BY SETTING ß=3.4, THE WEIBULL APPROXIMATES THE NORMAL.
RT = E –(T/Ө)ß
WHERE ß IS THE WEIBULL SLOPE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc31
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
OC CURVE CONSTRUCTION
STEPS:
1. ASSUME VALUES FOR THE MEAN LIFE Ө.
2. THESE VALUES ARE CONVERTED TO THE FAILURE RATE, Λ =1/Ө.
3. CALCULATE THE EXPECTED AVERAGE NUMBER OF FAILURES NTΛ.
4. FROM TABLE C OF THE APPENDIX USING NTΛ AND C VALUE, GET
PA.
5. SEE TABLE 11-1
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc32
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
OC Curve for the Sampling Plan, n = 16, T = 600 h, c = 2, and r = 3
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc33
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE AND RELIABILITY TESTING PLANS
TYPE OF TESTS:
FAILURE-TERMINATED: THESE LIFE-TEST SAMPLE PLANS ARE
TERMINATED WHEN A PREASSIGNED NUMBER OF FAILURES
OCCURS TO THE SAMPLE.
TIME-TERMINATED: THIS LIFE-TEST SAMPLING PLAN IS
TERMINATED WHEN THE SAMPLE OBTAINS A PREDETERMINED
TEST TIME.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc34
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE AND RELIABILITY TESTING PLANS
TYPE OF TESTS CONT’D.:
SEQUENTIAL: A THIRD TYPE OF LIFE-TESTING PLAN IS
A SEQUENTIAL LIFE-TEST SAMPLING PLAN WHEREBY
NEITHER THE NUMBER OF FAILURES NOR THE TIME
REQUIRED TO REACH A DECISION ARE FIXED IN
ADVANCE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc35
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE AND RELIABILITY TESTING PLANS
TESTS ARE BASED ON ONE OR MORE OF THE FOLLOWING
CHARACTERISTICS:
MEAN LIFE: THE AVERAGE LIFE OF THE PRODUCT.
FAILURE RATE: THE PERCENTAGE OF FAILURES PER UNIT TIME OR
NUMBER OF CYCLES.
HAZARD RATE: THE INSTANTANEOUS FAILURE RATE AT A SPECIFIED
TIME.
RELIABLE LIFE: THE LIFE BEYOND WHICH SOME SPECIFIED PORTION OF
THE ITEMS IN THE LOT WILL SURVIVE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc36
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
LIFE AND RELIABILITY TESTING PLANS
TEST ARE BASED ON ONE OR MORE OF THE FOLLOWING
CHARACTERISTICS CONT’D.:
HAZARD RATE: THE INSTANTANEOUS FAILURE RATE AT A
SPECIFIED TIME.
RELIABLE LIFE: THE LIFE BEYOND WHICH SOME SPECIFIED
PORTION OF THE ITEMS IN THE LOT WILL SURVIVE.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc37
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
THANK YOU
38
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
