Practical reliability

1
Introduction to Reliability
Engineering

WHAT IS RELIABILITY ENGINEERING?

No one disputes the need for articles to be reliable. The average consumer is acutely
aware of the problem of less than perfect reliability in domestic products such as TV
sets and automobiles. Organizations such as airlines, the military and public utilities
are aware of the costs of unreliability. Manufacturers often suffer high costs of
failure under warranty. Argument and misunderstanding begin when we try to
quantify reliability values, or try to assign financial or other benefit values to levels
of reliability.
The simplest, purely producer-oriented or inspectors' view of reliability is that in
which a product is assessed against a specification or set of attributes, and when
passed is delivered to the customer. The customer, having accepted the product,
accepts that it might fail at some future time. This simple approach is often coupled
with a warranty, or the customer may have some protection in law, so that he may
claim redress for failures occurring within a stated or reasonable time. However, this
approach provides no measure of quality over a period of time, particularly outside
a warranty period. Even within a warranty period, the customer usually has no
grounds for further action if the product fails once, twice or several times, provided
that the manufacturer repairs the product as promised each time. If it fails often, the
manufacturer will suffer high warranty costs, and the customers will suffer incon-
venience. Outside the warranty period, only the customer suffers. In any case, the
manufacturer will also probably incur a loss of reputation, possibly affecting future
business.
We therefore come to the need for a time-based concept of quality. The inspectors'
concept is not time-dependent. The product either passes a given test or it fails. On
the other hand, reliability is usually concerned with failures in the time domain. This
distinction marks the difference between traditional quality control and reliability
engineering.
Whether failures occur or not and their times to occurrence, can seldom be
forecast accurately. Reliability is therefore an aspect of engineering uncertainty.
Whether an item will work for a particular period is a question which can be
answered as a probability. This results in the usual engineering definition of reliabil-
ity as:
1

2 Introduction to Reliability Engineering

The probability that an item will perform a required function without failure under stated
conditions for a stated period of time.
Reliability can also be expressed as the number of failures over a period.
Durability is a particular aspect of reliability, related to the ability of an item to
withstand the effects of time (or of distance travelled, operating cycles, etc.) depend-
ent mechanisms such as fatigue, wear, corrosion, electrical parameter change, etc.
Durability is usually expressed as a minimum time before the occurrence of wearout
failures.
The objectives of reliability engineering, in the order of priority, are:

1. To apply engineering knowledge and specialist techniques to prevent or to
reduce the likelihood or frequency of failures.
2. To identify and correct the causes of failures that do occur, despite the efforts to
prevent them.
3. To determine ways of coping with failures that do occur, if their causes have not
been corrected.
4. To apply methods for estimating the likely reliability of new designs, and for
analysing reliability data.

The reason for the priority emphasis is that it is by far the most effective way of
working, in terms of minimizing costs and generating reliable products.
The primary skills that are required, therefore, are the ability to understand and
anticipate the possible causes of failures, and knowledge of how to prevent them. It
is also necessary to have knowledge of the methods that can be used for analysing
designs and data. The primary skills are nothing more than good engineering
knowledge and experience, so reliability engineering is the first and foremost the
application of good engineering, in the widest sense, during design, development,
manufacture and service.
Mathematical and statistical methods can be used for quantifying reliability
(prediction, measurement) and for analysing reliability data. The basic methods
are described in Chapter 2, to provide an introduction for some of the applications
described subsequently. However, because of the high levels of uncertainty in-
volved these can seldom be applied with the kind of precision and credibility that
engineers are accustomed to when dealing with most other problems. In practice the
uncertainty is often in orders of magnitude. Therefore the role of mathematical and
statistical methods in reliability engineering is limited, and appreciation of the
uncertainty is important in order to minimise the chances of performing inappropri-
ate analysis and of generating misleading results. Mathematical and statistical
methods can make valuable contributions in appropriate circumstances, but prac-
tical engineering must take precedence in determining the causes of problems and
their solutions. Unfortunately not all reliability training, literature and practice
reflect this reality.
Over-riding all of these aspects, though, is the management of the reliability
engineering effort. Since reliability (and very often also safety) is such a critical
parameter of most modern engineering products, and since failures are generated
primarily by the people involved (designers, test engineers, manufacturing, sup-
pliers, maintainers, users), it can be maximised only by an integrated effort that

Why teach Reliability Engineering? 3

encompasses training, teamwork, discipline, and application of the most appropri-
ate methods. Reliability engineering ``specialists'' cannot make this happen. They
can provide support, training and tools, but only managers can organise, motivate,
lead and provide the resources. Reliability engineering is, ultimately, effective
management of engineering.

WHY TEACH RELIABILITY ENGINEERING?

Engineering education is traditionally concerned with teaching how manufactured
products work. The ways in which products fail, the effects of failure and aspects of
design, manufacture, maintenance and use which affect the likelihood of failure are
not usually taught, mainly because it is necessary to understand how a product
works before considering ways in which it might fail. For many products the
tendency to approach the failed state is analogous to entropy. The engineer's
tasks are to design and maintain the product so that the failed state is deferred. In
these tasks he faces the problems inherent in the variability of engineering materials,
processes and applications. Engineering education is basically deterministic,
and does not usually pay sufficient attention to variability. Yet variability and
chance play a vital role in determining the reliability of most products. Basic
parameters like mass, dimensions, friction coefficients, strengths and stresses are
never absolute, but are in practice subject to variability due to process and materials
variations, human factors and applications. Some parameters also vary with time.
Understanding the laws of chance and the causes and effects of variability is
therefore necessary for the creation of reliable products and for the solution of
problems of unreliability.
However, there are practical problems in applying statistical knowledge to engin-
eering problems. These problems have probably deterred engineers in the past from
using statistical methods, and texts on reliability engineering and mathematics have
generally stressed the theoretical aspects without providing guidance on their
practical application. To be helpful a theoretical basis must be credible, and statis-
tical methods which work well for insurance actuaries, market researchers or agri-
cultural experimenters may not work as well for engineers. This is not because the
theory is wrong, but because engineers usually have to cope with much greater
degrees of uncertainty, mainly due to human factors in production and use.
Some highly reliable products are produced by design and manufacturing teams
who practise the traditional virtues of reliance on experience and maintenance of
high quality. They do not see reliability engineering as a subject requiring specialist
consideration, and a book such as this would teach them little that they did not
already practise in creating their reliable products. Engineers and managers might
therefore regard a specialist reliability discipline with scepticism. However, many
pressures now challenge the effectiveness of the traditional approaches. Competi-
tion, the pressure of schedules and deadlines, the cost of failures, the rapid evolution
of new materials, methods and complex systems, the need to reduce product costs,
and safety considerations all increase the risks of product development. Figure 1.1
shows the pressures that lead to the overall perception of risk. Reliability engineer-
ing has developed in response to the need to control these risks.

Competition Market pressure

Management emphasis
Safety

Perceived
Customer requirements
risk
Warranty and
service costs Legal, statutory

Public liability Development risks

Figure 1.1 Perception of risk

Later chapters will show how reliability engineering methods can be applied to
design, development and management to control the level of risk. The extent
to which the methods are applicable must be decided for each project and for each
design area. They must not replace normal good practice, such as safe design
for components subject to cyclic loading, or application guidelines for electronic
components. They should be used to supplement good practice. However, there
are times when new risks are being taken, and the normal rules and guidelines are
inadequate or do not apply. Sometimes we take risks unwittingly, when we assume
that we can extrapolate safely from our present knowledge. Designers and managers
are often overoptimistic or are reluctant to point out risks about which they are
unsure.
It is for these reasons that an understanding of reliability engineering principles
and methods is now an essential ingredient of modern engineering.

WHY DO ENGINEERING ITEMS FAIL?

There are many reasons why a product might fail. Knowing, as far as is practicable, the
potential causes of failures is fundamental to preventing them. It is rarely practicable to
anticipate all of the causes, so it is also necessary to take account of the uncertainty
involved. The reliability engineering effort, during design, development and in
manufacture and service should address all of the anticipated and possibly unantici-
pated causes of failure, to ensure that their occurrence is prevented or minimized.

The main reasons why failures occur are:

1. The design might be inherently incapable. It might be too weak, consume too much
power, suffer resonance at the wrong frequency, etc. The list of possible reasons is
endless, and every design problem presents the potential for errors, omissions,
and oversights. The more complex the design or difficult the problems to be
overcome, the greater is this potential.

Why do Engineering Items Fail? 5

2. The item might be overstressed in some way. If the stress applied exceeds the
strength then failure will occur. An electronic component will fail if the applied
electrical stress (voltage, current) exceeds the ability to withstand it, and a
mechanical strut will buckle if the compression stress applied exceeds the buck-
ling strength. Overstress failures such as these do happen, but fortunately not
very often, since designers provide margins of safety. Electronic component
specifications state the maximum rated conditions of application, and circuit
designers take care that these rated values are not exceeded in service. In most
cases they will in fact do what they can to ensure that the in-service worst case
stresses remain below the rated stress values: this is called `de-rating'. Mechanical
designers work in the same way: they know the properties of the materials being
used (e.g. ultimate tensile strength) and they ensure that there is an adequate
margin between the strength of the component and the maximum applied stress.
However, it might not be possible to provide protection against every possible
stress application. If a 110 V appliance is connected to a 240 V supply it will
probably fail, and a tyre will burst if sufficiently over-inflated.
3. Failures might be caused by variation. In the situations described above the values
of strength and load are fixed and known. If the known strength always exceeds
the known load, as shown in Fig. 1.2, then failure will not occur. However, in
most cases, there will be some uncertainty about both. The actual strength values
of any population of components will vary: there will be some that are relatively
strong, others that are relatively weak, but most will be of nearly average
strength. Also, the loads applied will be variable. Figure 1.3 shows this type of
situation. As before, failure will not occur so long as the applied load does not
exceed the strength. However, if there is an overlap between the distributions of
load and strength, and a load value in the high tail of the load distribution
is applied to an item in the weak tail of the strength distribution so that
there is overlap or interference between the distributions (Fig. 1.4), then failure
will occur. We will discuss load and strength interference in more detail in
Chapter 4.
4. Failures can be caused by wearout. We will use this term to include any mechan-
ism or process that causes an item that is sufficiently strong at the start of its life to

1
Probability

0
L S
Load − strength

Figure 1.2 Load±strengthÐdiscrete values


1

Probability
0
L S
Load − strength

Figure 1.3 Load±strengthÐdistributed values

1
Probability

0
L S
Load − strength

Figure 1.4 Load±strengthÐinterfering distributions

become weaker with age. Well-known examples of such processes are materi-
alfatigue, wear between surfaces in moving contact, corrosion, insulation deteri-
oration, and the wearout mechanisms of light bulbs and fluorescent tubes. Figure
1.5 illustrates this kind of situation. Initially the strength is adequate to withstand
the applied loads, but as weakening occurs over time the strength decreases. In
every case the average value falls and the spread of the strength distribution
widens. This is a major reason why it is so difficult to provide accurate predic-
tions of the lives of such items.
5. Failures can be caused by other time-dependent mechanisms. Battery run-down,
creep caused by simultaneous high temperature and tensile stress, as in turbine
discs and fine solder joints, and progressive drift of electronic component param-
eter values are examples of such mechanisms.
6. Failures can be caused by sneaks. A sneak is a condition in which the system
does not work properly even though every part does. For example, an electronic
system might be designed in such a way that under certain conditions incorrect
operation occurs. The fatal fire in the Apollo spacecraft crew capsule was caused in

Probabilistic Reliability 7

Strength

Load

tЈ Time/load cycles
Log scale

Figure 1.5 Time-dependent load and strength variation

this way: the circuit design ensured that an electrical short circuit would occur
when a particular sequence was performed by the crew. Sneaks can also occur in
software designs.
7. Failures can be caused by errors, such as incorrect specifications, designs or
software coding, by faulty assembly or test, by inadequate or incorrect mainten-
ance, or by incorrect use. The actual failure mechanisms that result might include
most of the list above.
8. There are many other potential causes of failure. Gears might be noisy, oil seals
might leak, display screens might flicker, operating instructions might be wrong
or ambiguous, electronic systems might suffer from electromagnetic interfer-
ence, etc.

Failures have many different causes and effects, and there are also different
perceptions of what kinds of events might be classified as failures. The burning
O-ring seals on the Space Shuttle booster rockets were not classed as failures, until
the ill-fated launch of Challenger. We also know that all failures, in principle and
almost always in practice, can be prevented.

PROBABILISTIC RELIABILITY

The concept of reliability as a probability means that any attempt to quantify it must
involve the use of statistical methods. An understanding of statistics as applicable to
reliability engineering is therefore a necessary basis for progress, except for the
special cases when reliability is perfect (we know the item will never fail) or it is
zero (the item will never work). For practical purposes a hammer might be 100 per
cent reliable if used for driving tacks, but would have a zero reliability if used to stop
a train. If used to break rocks, its reliability for a given period might have some value
between 0 and 100 per cent. In engineering we try to ensure 100 per cent reliability,


but our experience tells us that we do not always succeed. Therefore reliability
statistics are usually concerned with probability values which are very high (or
very low: the probability that a failure does occur, which is 1 ± reliability). Quanti-
fying such numbers brings increased uncertainty, since we need correspondingly
more information. Other sources of uncertainty are introduced because reliability is
often about people who make and people who use the product, and because of the
widely varying environments in which typical products might operate.
Further uncertainty, often of a subjective nature, is introduced when engineers
begin to discuss failures. Should a failure be counted if it was due to an error that is
hoped will not be repeated? If design action is taken to reduce the risk of one type of
failure, how can we quantify our trust in the designer's success? Was the machine
under test typical of the population of machines?
Reliability is quantified in other ways. We can specify a reliability as the mean
number of failures in a given time (failure rate), or as the mean time between failures
(MTBF) for items which are repaired and returned to use, or as the mean time to
failure (MTTF) for items which are not repaired. For repaired items, it is often
assumed that failures occur at a constant rate, in which case the failure rate
l ˆ (MTBF)À1 . However, this is only a special case, valuable because it is often
true and because it is easy to understand.
The application and interpretation of statistics in reliability are less straightfor-
ward than in, say, public opinion polls or measurement of human variations such as
IQ or height. In these applications, most interest is centred around the behaviour of
the larger part of the population or sample, variation is not very large and data are
plentiful. In reliability we are concerned with the behaviour in the extreme tails of
distributions and possibly unlikely combinations of load and strength, where vari-
ability is often hard to quantify and data are expensive.
Further difficulties arise in application of statistical theory to reliability engineer-
ing, owing to the fact that variation is often a function of time or of time-related
factors such as operating cycles, diurnal or seasonal cycles, maintenance periods,
etc. Engineering, unlike most fields of knowledge, is primarily concerned with
change, hopefully, but not always, for the better. Therefore the reliability data
from any past situation cannot be used to make credible forecasts of the future
behaviour, without taking into account non-statistical factors such as design
changes, maintainer training, and even imponderables such as unforeseeable pro-
duction or service problems. The statistician working in reliability engineering
needs to be aware of these realities.
Chapter 2 provides the statistical basis of reliability engineering, but it must
always be remembered that quality and reliability data contain many sources of
uncertainty and variability which cannot be rigorously quantified. It is also import-
ant to appreciate that failures and their causes are by no means always clear-cut and
unambiguous. They are often open to interpretation and argument. They also differ
in terms of importance (cost, safety, other effects). Therefore we must be careful not
to apply conventional scientific, deterministic thinking to the interpretation of
failures. For example, a mere count of total reported failures of a product is seldom
useful or revealing. It tells us nothing about causes or consequences, and therefore
nothing about how to improve the situation. This contrasts with a statement of a
physical attribute such as weight or power consumption, which is usually unam-

Repairable and Non-Repairable Items 9

biguous and complete. Nevertheless, it is necessary to derive values for decision-
making, so the mathematics are essential. The important point is that the reliability
engineer or manager is not, like an insurance actuary, a powerless observer of his
statistics. Statistical derivations of reliability are not a guarantee of results, and these
results can be significantly affected by actions taken by quality and reliability
engineers and managers.

REPAIRABLE AND NON-REPAIRABLE ITEMS

It is important to distinguish between repairable and non-repairable items when
predicting or measuring reliability.
For a non-repairable item such as a light bulb, a transistor, a rocket motor or an
unmanned spacecraft, reliability is the survival probability over the item's expected
life, or for a period during its life, when only one failure can occur. During the item's life
the instantaneous probability of the first and only failure is called the hazard rate. Life
values such as the mean life or mean time to failure (MTTF), or the expected life by
which a certain percentage might have failed (say 10 per cent.), are other reliability
characteristics that can be used. Note that non-repairable items may be individual
parts (light bulbs, transistors) or systems comprised of many parts (spacecraft,
microprocessors). When a part fails in a non-repairable system, the system fails
(usually) and system reliability is, therefore, a function of the time to the first part
failure.
For items which are repaired when they fail, reliability is the probability that
failure will not occur in the period of interest, when more than one failure can occur. It
can also be expressed as the failure rate or the rate of occurrence of failures (ROCOF).
However, the failure rate expresses the instantaneous probability of failure per unit
time, when several failures can occur in a time continuum. It is unfortunate that the
expression failure rate has been applied loosely in the past to non-repairable items.
What is really meant is that, in a repairable system which contains a part type, the
part will contribute by that amount to the system failure rate. The part, being non-
repairable, cannot have a failure rate. This distinction does not always have practical
significance, but it can, as will be shown later. The expression ROCOF is sometimes
used to stress this point.
Repairable system reliability can also be characterized by the mean time between
failures (MTBF), but only under the particular condition of a constant failure rate.
We are also concerned with the availability of repairable items, since repair
takes time. Availability is affected by the rate of occurrence of failures (failure rate)
and by maintenance time. Maintenance can be corrective (i.e. repair) or preventive
(to reduce the likelihood of failure, e.g. lubrication). We therefore need to under-
stand the relationship between reliability and maintenance, and how both reliability
and maintainability can affect availability.
Sometimes an item may be considered as both repairable and non-repairable. For
example, a missile is a repairable system whilst it is in store and subjected to
scheduled tests, but it becomes a non-repairable system when it is launched. Reli-
ability analysis of such systems must take account of these separate states. Repair-
ability might also be determined by other considerations. For example, whether a


TV electronics board is treated as a repairable item or not will depend upon the
cost of repair. An engine or vehicle will be treated as repairable only up to a certain
age.
Repairable system reliability data analysis is covered in Chapter 12 and availabil-
ity and maintainability in Chapter 14.

THE PATTERN OF FAILURES WITH TIME
(NON-REPAIRABLE ITEMS)

There are three basic ways in which the pattern of failures can change with time. The
hazard rate may be decreasing, increasing or constant. We can tell much about
the causes of failure and about the reliability of the item by appreciating the way
the hazard rate behaves in time.
A constant hazard rate is characteristic of failures which are caused by the
application of loads in excess of the design strength, at a constant average rate.
For example, overstress failures due to accidental or transient circuit overload, or
maintenance-induced failures of mechanical equipment, typically occur randomly
and at a generally constant rate.
Material fatigue brought about by strength deterioration due to cyclic loading is a
failure mode which does not occur for a finite time, and then exhibits an increasing
probability of occurrence.
Decreasing hazard rates are observed in items which become less likely to fail
as their survival time increases. This is often observed in electronic equipment
and parts. `Burn-in' of electronic parts is a good example of the way in which
knowledge of a decreasing hazard rate is used to generate an improvement in
reliability. The parts are operated under failure-provoking stress conditions for a
time before delivery. As substandard parts fail and are rejected the hazard rate
decreases and the surviving population is more reliable.
The combined effect generates the so-called bathtub curve (Figure 1.6). This shows
an initial decreasing hazard rate or infant mortality period, an intermediate useful life
period and a final wearout period. Death is a good analogy to failure of a non-
repairable system, and the bathtub curve model is similar to actuarial statistical
models.

THE PATTERN OF FAILURES WITH TIME (REPAIRABLE ITEMS)

The failure rates (or ROCOF) of repairable items can also vary with time, and
important implications can be derived from these trends.
A constant failure rate (CFR) is indicative of externally induced failures, as in the
constant hazard rate situation for non-repairable items. A CFR is also typical of
complex systems subject to repair and overhaul, where different parts exhibit
different patterns of failure with time and parts have different ages since repair or
replacement. Repairable systems can show a decreasing failure rate (DFR) when
reliability is improved by progressive repair, as defective parts which fail relatively

The Development of Reliability Engineering 11

Wearout failures
Total hazard rate 'bathtub' of 'good' items
Hazard rate

Externally
induced
Failure of weak items failures

Infant mortality Useful life Wearout
Time

Figure 1.6 The `bathtub' curve

early are replaced by good parts. `Burn in' is applied to electronic systems, as well as
to parts, for this purpose.
An increasing failure rate (IFR) occurs in repairable systems when wearout failure
modes of parts begin to predominate.
The pattern of failures with time of repairable systems can also be illustrated by
use of the bathtub curve (Fig. 1.6), but with the failure rate (ROCOF) plotted against
age instead of the hazard rate.
The statistical treatment of failure data is covered in Chapters 2 and 3.

THE DEVELOPMENT OF RELIABILITY ENGINEERING

Reliability engineering, as a separate engineering discipline, originated in the
United States during the 1950s. The increasing complexity of military electronic
systems was generating failure rates which resulted in greatly reduced availability
and increased costs. Solid state electronics technology offered long term hope, but
conversely miniaturization was to lead to proportionately greater complexity, which
offset the reliability improvements expected. The gathering pace of electronic device
technology meant that the developers of new military systems were making increas-
ing use of large numbers of new component types, involving new manufacturing
processes, with the inevitable consequences of low reliability. The users of such
equipment were also finding that the problems of diagnosing and repairing the new
complex equipment were seriously affecting its availability for use, and the costs of
spares, training and other logistics support were becoming excessive. Against this
background the US Department of Defense and the electronics industry jointly set
up the Advisory Group on Reliability of Electronic Equipment (AGREE) in 1952. The
AGREE report concluded that, to break out of the spiral of increasing development
and ownership costs due to low reliability, disciplines must be laid down as integral
activities in the development cycle for electronic equipment. The report laid particu-
lar stress on the need for new equipments to be tested for several thousand hours in


high stress cyclical environments including high and low temperatures, vibration
and switching, in order to discover the majority of weak areas in a design at an early
enough stage to enable them to be corrected before production commenced. Until
that time, environmental tests of tens of hours duration had been considered
adequate to prove the suitability of a design. The report also recommended that
formal demonstrations of reliability, in terms of statistical confidence that a specified
MTBF had been exceeded, be instituted as a condition for acceptance of equipment
by the procuring agency. A large part of the report was devoted to providing
detailed test plans for various levels of statistical confidence and environmental
conditions.
The AGREE report was accepted by the Department of Defense, and AGREE
testing quickly became a standard procedure. Companies which invested in
the expensive environmental test equipment necessary soon found that they could
attain levels of reliability far higher than by traditional methods. It was evident that
designers, particularly those working at the fringes of advanced technology, could
not be expected to produce highly reliable equipment without it being subjected to
a test regime which would show up weaknesses. Complex systems and
the components used in them included too many variables and interactions
for the human designer to cope with infallibly, and even the most careful design re
views and disciplines could not provide sufficient protection. Consequently it was
necessary to make the product speak for itself, by causing it to fail, and then to
eliminate the weaknesses that caused the failures. The Department of Defense
(DOD) reissued the AGREE report on testing as US Military Standard (MIL-STD)
781, Reliability Qualification and Production Approval Tests.
Meanwhile the revolution in electronic device technology continued, led
by integrated microcircuitry. Increased emphasis was now placed on improving
the quality of devices fitted to production equipments. Screening techniques,
in which devices are temperature cycled, vibrated, centrifuged, operated at electrical
overstress and otherwise abused, were introduced in place of the traditional
sampling techniques. With component populations on even single printed circuit
boards becoming so large, sampling no longer provided sufficient protection
against the production of defective equipment. These techniques were formalized
in military standards covering the full range of electronic components. Compon-
ents produced to these standards were called `Hi-rel' components. Specifications
and test systems for electronic components, based upon the US Military
Standards, have been developed in the United Kingdom and in continental
Europe, and internationally through the International Electrotechnical Commission
(IEC).
Engineering reliability effort in the United States developed quickly, and the
AGREE and reliability programme concepts were adopted by NASA and many
other major suppliers and purchasers of high technology equipment. In 1965 the
DOD issued MIL-STD-785±Reliability Programs for Systems and Equipment. This docu-
ment made mandatory the integration of a programme of reliability engineering
activities with the traditional engineering activities of design, development and
production, as it was by then realized that such an integrated programme was the
only way to ensure that potential reliability problems would be detected and
eliminated at the earliest, and therefore the cheapest, stage in the development

The Development of Reliability Engineering 13

cycle. Much written work appeared on the cost-benefit of higher reliability, to show
that effort and resources expended during early development and during produc-
tion testing, plus the imposition of demonstrations of specified levels of reliability to
MIL-STD-781, led to reductions in in-service costs which more than repaid the
reliability programme expenditure. The concept of life cycle costs (LCC), or whole
life costs, was introduced.
In the United Kingdom, Defence Standard 00±40, The Management of Reliability and
Maintainability was issued in 1981. The British Standards Institution issued BS 5760Ð
Guide on Reliability of Systems, Equipments and Components.
Starting in the early 1980s, the reliability of new Japanese industrial and commer-
cial products took Western competitors by surprise. Items such as automobiles,
electronic components and systems, and machine tools achieved levels of reliability
far in excess of previous experience. These products were also less expensive and
often boasted superior features and performance. The `Japanese quality revolution'
had been driven by the lessons taught by American teachers brought in to help
Japan's post-war recovery. The two that stand out were J. R. Juran and W. Edwards
Deming (Reference 9), who taught the principles of `total quality management'
(TQM) and continuous improvement. Japanese pioneers, particularly K. Ishikawa,
also contributed. These ideas were all firmly rooted in the teaching of the American
writer on management, Peter Drucker (Reference 10), that people work most effect-
ively when they are given the knowledge and authority to identify and implement
improvements, rather than being expected to work to methods dictated by `man-
agement'.
These ideas proved to be much more effective than those prescribed in documents
such as the standards mentioned above. They led to great increases in productivity
and quality, and thus in market penetration, as Drucker had predicted. Many
Western companies followed the new path that had been blazed and also made
great improvements. However, the results of the Japanese head start are still appar-
ent today in showrooms, shops, offices and factories.
Today we expect very high reliability in all of the products we use. Modern
electronic components and systems seldom fail, and complex systems such as
domestic equipment, automobiles and aircraft are far more reliable than the much
simpler systems of twenty or thirty years ago. These improvements have been
generated primarily by the application of the methods of TQM and continuous
improvement. Many of the methods that had been standardized in the West have
been shown to be less effective.
Another aspect of reliability thinking that has developed is the application of
statistical methods. Since reliability can be expressed as a probability, and is affected
by variation, in principle these methods are applicable. They form the basis of most
teaching and literature on the subject. However, variation in engineering is usually
of such an uncertain nature that refined mathematical and quantitative techn-
iques can be inappropriate and misleading. This aspect will be discussed in later
chapters.
Quality control (QC) of manufacturing processes obviously makes an essential
contribution to the reliability of a product. In the chapters on manufacturing
(Chapter 13) and reliability management (Chapter 15), the need to consider QC as
an integral part of an overall reliability programme will be emphasized.


COURSES, CONFERENCES AND LITERATURE

Reliability engineering and management are now taught in engineering courses at a
large number of universities, colleges and polytechnics, and on specialist short
courses.
Conferences on general and specific reliability engineering and management
topics have been held regularly in the United States since the 1960s and in Europe
and elsewhere since the 1970s. The best known is the annual US Reliability and
Maintainability Symposium (RAMS), sponsored by most of the important engineer-
ing associations and institutions in the United States. These conference proceedings
contain much useful information and are often cited.
Journals on reliability have also appeared; some are referenced at the end of this
chapter. Several books have been published on the subjects of reliability engineering
and management; some of these are referenced at the end of other chapters.
Much of the reliability literature has tended to emphasize the mathematical and
analytical aspects of the subject, with the result that reliability engineering is often
considered by designers and others to be a rather esoteric subject. This is unfortunate,
since it creates barriers to communication. More recently there has been a trend to
emphasize the more practical aspects and to integrate reliability work into the overall
management and engineering process. These aspects are covered in later chapters.

ORGANIZATIONS INVOLVED IN RELIABILITY WORK

Several organizations have been created to develop policies and methods in reliabil-
ity engineering and to undertake research and training. A summary of these is given
in the book's homepage (see Preface to the Fourth Edition, page xxi).

RELIABILITY AS AN EFFECTIVENESS PARAMETER

With the increasing cost and complexity of many modern systems, the importance of
reliability as an effectiveness parameter, which should be specified and paid for, has
become apparent. For example, a radar station, a process plant or an airliner must be
available when required, and the cost of non-availability, particularly if it is un-
scheduled, can be very high. In the weapons field, if an anti-aircraft missile has a less
than 100 per cent probability of functioning correctly throughout its engagement
sequence, operational planners must consider deploying the appropriate extra
quantity to provide the required level of defence. The Apollo project second stage
rocket was powered by six rocket motors; any five would have provided sufficient
impulse, but an additional motor was specified to cater for a possible failure of one.
As it happened there were no failures, and every launch utilized an `unnecessary'
motor. These considerations apply equally to less complex systems, e.g. vending
and copying machines, even if the failure costs are less dramatic in absolute terms.
As an effectiveness parameter, reliability can be `traded off' against other param-
eters. Reliability generally affects availability, and in this context maintainability is
also relevant. Reliability and maintainability are often related to availability by the
formula:

Reliability Programme Activities 15
MTBF
Availability ˆ
MTBF ‡ MTTR

where MTTR is the mean time to repair. This is the simplest steady-state situation. It
is clear that availability improvements can be achieved by improving either MTBF
or MTTR. For example, automatic built-in test equipment can greatly reduce diag-
nostic times for electronic equipment, at a cost of a slight reduction in overall
reliability and an increase in unit costs. Many other parameters can be considered
in trade-offs, such as weight, redundancy, cost of materials, parts and processes, or
reduction in performance.
The greatest difficulty in estimating relationships for reliability trade-offs derives
from the fact that, whereas it is possible to estimate quite accurately such factors
as the cost and weight penalties of built-in test equipment, the cost of materials and
components, or the worth of a measurable performance parameter, the effect on
reliability cannot generally be forecast accurately, and reliability measurements can
at best be made only within statistical limits imposed by the amount of data
available. Even for quite simple systems, quantifying the separate contributions to
a measured change in reliability due to different causes is a very uncertain exercise.
Consequently, while there is considerable literature on the effectiveness of various
reliability methods, few generally applicable estimating relationships are available.
Selection of trade-offs must therefore be very much a matter of experience of similar
projects in the knowledge that wide margins of error can exist.

RELIABILITY PROGRAMME ACTIVITIES

What, then, are the actions that managers and engineers can take to influence
reliability? One obvious activity already mentioned is quality control, the whole
range of functions designed to ensure that delivered products are compliant with
the design. For many products, QC is sufficient to ensure high reliability, and we
would not expect a match factory or a company mass-producing simple diecastings
to employ reliability staff. In such cases the designs are simple and well proven, the
environments in which the products will operate are well understood and the very
occasional failure has no significant financial or operational effect. QC, together with
craftsmanship, can provide adequate assurance for simple products or when the
risks are known to be very low. Risks are low when safety margins can be made very
large, as in most structural engineering. Reliability engineering disciplines may
justifiably be absent in many types of product development and manufacture. QC
disciplines are, however, essential elements of any integrated reliability programme.
A formal reliability programme is necessary whenever the risks or costs of failure
are not low. We have already seen how reliability engineering developed as a result
of the high costs of unreliability of military equipment, and later in commer-
cial application. Risks of failure usually increase in proportion to the number of
components in a system, so reliability programmes are required for any product
whose complexity leads to an appreciable risk.
An effective reliability programme should be based on the conventional wisdom
of responsibility and authority being vested in one person. Let us call him or her the


reliability programme manager. The responsibility must relate to a defined object-
ive, which may be a maximum warranty cost figure, an MTBF to be demonstrated or
a requirement that failure will not occur. Having an objective and the authority, how
does the reliability programme manager set about his or her task, faced as he or she
is with a responsibility based on uncertainties? This question will be answered in
detail in subsequent chapters, but a brief outline is given below.
The reliability programme must begin at the earliest, conceptual phase of the
project. It is at this stage that fundamental decisions are made, which can signifi-
cantly affect reliability. These are decisions related to the risks involved in the
specification (design, complexity, producibility, etc.), development time-scale, re-
sources applied to evaluation and test, skills available, and other factors.
The shorter the development time-scale, the more important is this need, particu-
larly if there will be few opportunities for an iterative approach. The activities
appropriate to this phase are an involvement in the assessment of these trade-offs
and the generation of reliability objectives. The reliability staff can perform these
functions effectively only if they are competent to contribute to the give-and-take
inherent in the trade-off negotiations, which may be conducted between designers,
production staff, marketing staff, customer representatives and finance staff.
As development proceeds from initial study to detail design, the reliability risks
are controlled by a formal, documented approach to the review of design and to the
imposition of design rules relating to components, materials and process selection,
de-rating policy, tolerancing, etc. The objectives at this stage are to ensure that
known good practices are applied, that deviations are detected and corrected, and
that areas of uncertainty are highlighted for further action. The programme con-
tinues through the initial hardware manufacturing and test stages, by planning and
executing tests to generate confidence in the design and by collecting, analysing and
acting upon test data. During production, QC activities ensure that the proven
design is repeated, and further testing may be applied to eliminate weak items
and to maintain confidence. The data collection, analysis and action process con-
tinues through the production and in-use phases. Throughout the product life cycle,
therefore, the reliability is assessed, first by initial predictions based upon past
experience in order to determine feasibility and to set objectives, then by refining
the predictions as detail design proceeds and subsequently by recording perform-
ance during the test, production and in-use phases. This performance is fed back to
generate corrective action, and to provide data and guidelines for future products.
The elements of a reliability programme are outlined in documents such as US
MIL-STD-785, UK Defence Standard 00±40 and British Standard 5760 (References
1±3). The activities are described fully in subsequent chapters.

RELIABILITY ECONOMICS AND MANAGEMENT

Obviously the reliability programme activities described can be expensive. Figure
1.7 is a commonly-described representation of the theoretical cost±benefit relation
ship of effort expended on reliability (or production quality) activities. However,
despite its intuitive appeal and frequent presentation in textbooks and teaching on
quality and reliability, this picture is misleading. Closer thought easily uncovers the

Reliability Economics and Management 17

Total costs
Quality/Reliability
Programme costs
Cost

Failure costs

Quality/Reliability 100%

Figure 1.7 Reliability and life cycle costs (traditional view)

error in the picture. Since less than perfect reliability is the result of failures, all of
which have causes, we should ask `what is the cost of preventing or correcting the
cause, compared with the cost of doing nothing?' When each potential or actual
cause is analysed in this way, it is nearly always apparent that total costs continue to
reduce indefinitely as reliability is improved. In other words, all effort on an
effective reliability programme represents an investment, usually with a large
payback over a short period. The only problem is that it is not easy to quantify the
effects of given reliability programme activities, such as a certain amount of testing,
on achieved reliability. However, experience shows clearly that the more realistic
picture is as shown in Fig. 1.8. It was W. E. Deming (Reference 9) who first explained
this relationship in his teaching on production quality. The truth has been clearly
demonstrated by the success of the companies that have wholeheartedly adopted
this teaching.
Achieving reliable designs and products requires a totally integrated approach, in-
cluding design, training, test, production, as well as the reliability programme activ-
ities. In such an approach it is difficult to separately identify and cost those activities
that are specifically devoted to reliability, as opposed to say performance or cost. The
integrated engineering approach places high requirements for judgement and engin-
eering knowledge on project managers and team members. Reliability specialists
must play their parts as members of the team.
There are three kinds of engineering product, from the perspective of failure
prevention:

1. Intrinsically reliable components, which are those that have high margins between
their strength and the stresses that could cause failure, and which do not wear out
within their practicable lifetimes. Such items include nearly all electronic com-
ponents (if properly applied), nearly all mechanical non-moving components,
and all correct software.


Total costs

Cost Development/Production costs

Failure costs

Quality/Reliability 100%

Figure 1.8 Reliability and life cycle costs (modern view)

2. Intrinsically unreliable components, which are those with low design margins or
which wear out, such as badly applied components, light bulbs, turbine blades,
parts that move in contact with others, like gears, bearings and power drive
belts, etc.
3. Systems which include many components and interfaces, like cars, dishwashers,
aircraft, etc, so that there are many possibilities for failures to occur, particularly
across interfaces (e.g. inadequate electrical overstress protection, vibration nodes
at weak points, electromagnetic interference, software that contains errors, etc.).

It is the task of design engineers to ensure that all components are correctly applied,
that margins are adequate (particularly in relation to the possible extreme values of
strength and stress, which are often variable), that wearout failure modes are
prevented during the expected life (by safe life design, maintenance, etc.), and that
system interfaces cannot lead to failure (due to interactions, tolerance mismatches,
etc.). Because achieving all this on any modern engineering product is a task that
challenges the capabilities of the very best engineering teams, it is almost certain that
aspects of the initial design will fall short of the `intrinsically reliable' criterion.
Therefore we must submit the design to analyses and tests in order to show not only
that it works, but also to show up the features that might lead to failures. When we
find out what these are we must redesign and re-test, until the final design is
considered to meet the criterion.
Then the product has to be manufactured. In principle, every one should be identi-
cal and correctly made. Of course this is not achievable, because of the inherent
variability of all manufacturing processes, whether performed by humans or by
machines. It is the task of the manufacturing people to understand and control
variation, and to implement inspections and tests that will identify non-conforming
product.

Bibliography 19

For many engineering products the quality of operation and maintenance also
influence reliability.
The essential points that arise from this brief and obvious discussion of failures
are that:

1. Failures are caused primarily by people (designers, suppliers, assemblers, users,
maintainers). Therefore the achievement of reliability is essentially a manage-
ment task, to ensure that the right people, skills, teams and other resources are
applied to prevent the creation of failures.
2. Reliability (and quality) are not separate specialist functions that can effectively
ensure the prevention of failures. They are the results of effective working by all
involved.
3. There is no fundamental limit to the extent to which failures can be prevented.
We can design and build for ever-increasing reliability.

Deming explained how, in the context of manufacturing quality, there is no point
at which further improvement leads to higher costs. This is, of course, even more
powerfully true when considered over the whole product life cycle, so that efforts to
ensure that designs are intrinsically reliable, by good design and effective develop-
ment testing, generate even higher pay-offs than improvements in production
quality. The `kaizen' (continuous improvement) principle is even more effective
when applied to up-front engineering.
The creation of reliable products is, therefore, primarily a management task. Guidance on
reliability programme management and costs is covered in Chapter 15.

BIBLIOGRAPHY

1. US MIL-STD-785: Reliability Programs for Systems and Equipment. National Technical Infor-
mation Service, Springfield, Virginia (suspended in 1976).
2. UK Defence Standard 00±40: The Management of Reliability and Maintainability. HMSO.
3. British Standard, BS 5760: Reliability of Systems, Equipments and Components. British Stand-
ards Institution, London.
4. US MIL-STD-721: Definitions of Effectiveness Terms for Reliability, Maintainability, Human
Factors and Safety. Available from the National Technical Information Service, Springfield,
Virginia.
5. British Standard, BS 4778: Glossary of Terms Used in Quality Assurance (including reliability
and maintainability). British Standards Institution, London.
6. Quality and Reliability Engineering International. Wiley (published quarterly).
7. IEEE Trans. Reliab. IEEE (published quarterly).
8. Proceedings of the US Reliability and Maintainability Symposia. American Society for Quality
and IEEE (published annually).
9. W. E. Deming, Out of the Crisis. MIT University Press, 1986 (originally published under the
title Quality, Productivity and Competitive Position).
10. P. F. Drucker, The Practice of Management. Heinemann (1995).


QUESTIONS

1. Define (a) failure rate, and (b) hazard rate. Explain their application to the reliability of
components and repairable systems. Discuss the plausibility of the `bathtub curve' in
both contexts.

2. (a) Explain the theory of component failures derived from the interaction of stress (or
load) and strength distributions. Explain how this theory relates to the behaviour
of the component hazard function.
(b) Discuss the validity of the `bathtub curve' when used to describe the failure
characteristics of non-repairable components.

3. What are the main objectives of a reliability engineering team working on an engin-
eering development project? Describe the important skills and experience that should
be available within the team.

4. Briefly list the most common basic causes of failures of engineering products.

5. It is sometimes claimed that increasing quality and reliability beyond levels that have
been achieved in the past is likely to be uneconomic, due to the costs of the actions that
would be necessary. Present the argument against this belief. Illustrate it with an
example from your own experience.

Practical reliability

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Practical reliability (20)

Recently uploaded (20)

Practical reliability