Reliability Blueprint PDF

ReliaSoft Corporation R&D Reports Blueprint for a Comprehensive Reliability Program
Blueprint for a
Comprehensive Reliability
Program
ReliaSoft Corporation R & D Reports

Author: Smith, Crawford
Last Revised on: February 3, 2003
Prepared by:
ReliaSoft Corporation
Phone: +1.520.886.0410
Fax: +1.520.886.0399
www.ReliaSoft.com
ReliaSoft Corporation Page 1

© 2000-2003. ReliaSoft Corporation. All Rights Reserved. Revised: 2/3/03
TABLE OF CONTENTS
1 Introduction ................................................................................................... 1
2 Foundations of Reliability ............................................................................ 2
2.1 Fostering a Culture of Reliability .......................................................................... 2
2.2 Product Mission ................................................................................................... 3
2.2.1 Reliability Specifications .............................................................................................3
2.3 Universal Failure Definitions ................................................................................ 4
3 Reliability Testing ......................................................................................... 6
3.1 Customer Usage Profiling .................................................................................... 6
3.2 Test Types ........................................................................................................... 6
3.2.1 Development Testing..................................................................................................7
3.2.2 Manufacturing Testing ................................................................................................8
4 Field Data..................................................................................................... 10
4.1 The “Disconnect” Between In-House and Field Data......................................... 10
4.2 Sales and Forecasting (Shipping) Data ............................................................. 10
4.3 Warranty Data.................................................................................................... 11
4.4 Field Service Data.............................................................................................. 11
4.5 Customer Support Data ..................................................................................... 12
4.6 Returned Parts/Failure Analysis Data................................................................ 12
5 Data Collection............................................................................................ 13
5.1 In-House Test Data Collection ........................................................................... 13
5.1.1 Test Log ....................................................................................................................14
5.1.2 Failure Log ................................................................................................................14
5.1.3 Service Log ...............................................................................................................15
5.2 Field Data Collection.......................................................................................... 16
5.2.1 ReliaSoft’s Dashboard ..............................................................................................16
6 Data Analysis and Reporting ..................................................................... 19
6.1 Non-Parametric Analysis ................................................................................... 19
6.1.1 Non-Parametric Reliability Analysis..........................................................................19
6.2 Parametric Analysis ........................................................................................... 21
6.2.1 Examples of Reporting for Parametric Data Analysis ..............................................22
6.2.1.1 Probability Plot...................................................................................................22
6.2.1.2 Reliability Function ............................................................................................23
6.2.1.3 Probability Density Function ..............................................................................23
6.2.1.4 Failure Rate Function ........................................................................................23
6.2.1.5 Life vs. Stress ....................................................................................................24
6.2.1.6 Likelihood Function............................................................................................24
6.2.1.7 Reliability Importance ........................................................................................25
6.2.1.8 Reliability Growth...............................................................................................25
7 Bringing It All Together .............................................................................. 26
7.1 Connecting Field and Lab Data ......................................................................... 26
7.2 Reliability Growth ............................................................................................... 26
7.3 Optimum Design Level Determination ............................................................... 27
7.4 Competitive Assessment ................................................................................... 28
7.5 Marketing and Advertising ................................................................................. 28

1 INTRODUCTION
At the highest level, the purpose of a reliability engineering program is to test and
report on the reliability of an organization's products. This information is then
used to assess the financial impact of the reliability of the products, and to
improve the overall product reliability and consequently the financial strength of
the organization.
Reliability assessment is based on the results of testing from in-house labs and
data pertaining to the performance results of the product in the field. The data
produced by these sources is to be utilized to accurately measure and improve
the reliability of the products being produced. This is particularly important as
market concerns drive a constant push for cost reduction. However, one must be
able to keep a perspective on "the big picture," instead of merely looking for the
quick fix. It is often the temptation to cut corners and save initial costs by using
cheaper parts or cutting testing programs. Unfortunately, cheaper parts are
usually less reliable, and inadequate testing programs can allow products with
undiscovered flaws to get out into the field. A quick savings in the short term by
the use of cheaper components or small test sample sizes will usually result in
higher long-term costs in the form of warranty costs, or loss of customer
confidence. The proper balance must be struck between reliability, customer
satisfaction, time to market, sales and features. Figure 1 illustrates this concept.
The polygon on the left represents a properly balanced project. The polygon on
the right represents a project in which reliability and customer satisfaction have
been sacrificed for the sake of sales and time to market.
t t
r ke Re r ke Re
a li ab a lia
M ility M bil
to to ity
e e
iT m iT m
ion
ion
Sale
fact
fact
Sale
s
Satis
Satis
s
Features Features
Figure 1 - Graphical Representation of Balanced and Unbalanced Projects.
As the figure above illustrates, a proper balance must be struck between product
reliability and the other aspects of the business, including time-to-market,
manufacturing costs, sales, product features, and customer satisfaction. Through
proper testing and analysis in the in-house testing labs, as well as collection of
adequate and meaningful data on a product's performance in the field, the
reliability of any product can be measured, tracked and improved, leading to a
balanced organization with a financially healthy outlook for the future.

2 FOUNDATIONS OF RELIABILITY
There are a certain number of "up-front" activities that need to be undertaken in
order to be able to successfully implement a reliability program. Most of these
activities are relatively simple, but they are vital to the design and implementation
of a reliability program. It is like the foundation of a house: relatively inexpensive
and largely forgotten once the major construction has begun, but vital to the
structure nonetheless. The reliability "foundation building" concepts need to be
addressed in order to put a strong reliability program in place.
2.1 FOSTERING A CULTURE OF RELIABILITY

The most important part of developing a reliability program is having a culture of
reliability in the organization. It is vital that everyone involved in the product's
production, from upper management to assembly personnel, understands that a
sound reliability program is necessary for the organization's success.
Achieving this culture of reliability may actually be more difficult than it seems, as
some organizations may not have the history or background that lends itself to
the support of a reliability program. This can be particularly true in situations
where the organization has had a niche market or little or no previous
competition with the products that it produces. In the past, the organization's
customers may have had to accept the reliability of the product, good or bad. As
a consequence, the organization may have developed a mentality that tends to
overlook the reliability of a product in favor of the "damn-the-torpedoes, full-
steam-ahead" method of product development. In this type of organization,
reliability engineering methods and practices tend to be viewed as superfluous or
even wasteful. "We don't need all of this reliability stuff, we'll just find the
problems and fix them," tends to be the attitude in these circumstances.
Unfortunately, this attitude often results in poorly-tested, unreliable products
being shipped to customers.
The first step in developing the necessary culture of reliability is to have the
support of the organization's top management. Without this, the implementation
of a reliability program will be a difficult and frustrating process. Adequate
education as to the benefits of a properly constructed reliability program will go a
long way towards building support in upper management for reliability techniques
and processes. Most important is the emphasis of the financial benefits that will
accrue from a good reliability program, particularly in the form of decreased
warranty costs and increased customer goodwill. This latter aspect of the
benefits of reliability engineering can sometimes be an elusive concept to
appreciate. An adage in the reliability field states that if customers are pleased
with a product, they will tell eight other people, but if they are dissatisfied, they
will tell twenty-two other people.1 While this anecdote is rather eye-opening, it
must be put in a financial context to have the full impact. It is possible to
construct a model that links reliability levels of a product with the probability of
repeat sales. It is therefore possible to calculate a loss of sales revenue based
on the unreliability of the product. This type of information is useful in educating
upper management on the financial importance of reliability. Once the upper
management has been adequately educated and is supportive of the
1
This adage is rather dated. Given the popularity of the Internet and the proliferation of newsgroups and
web sites dedicated to disgruntled customers and employees, this number is probably even higher than
twenty-two.

implementation of reliability concepts, it will be a great deal easier to go about

implementing those concepts.
However, one should not stop with upper management when it comes to
educating an organization on the benefits of a proposed reliability program. It is
vital to have the support and understanding of the rest of the organization as
well. Since the implementation of a reliability program will affect the day-to-day
activities of middle management, the engineers, and the technicians, it is also
necessary to convince these groups of the benefits of reliability-oriented
activities. It is important to demonstrate to them how these activities, which may
initially seem pointless or counterproductive to them, will ultimately benefit the
organization. For example, if test technicians have a good understanding of how
the test data are going to be put to use, they will be less likely to cut corners
while performing the test and recording the data. Overall, a reliability program
stands the greatest chance of success if everyone in the organization
understands the benefits and supports the techniques involved.
2.2 PRODUCT MISSION

The underlying concept in characterizing the reliability of a product involves the
concept of product mission (e.g., operate for 36 months, or complete 1000
cycles). A textbook definition of reliability is:
“The conditional probability, at a given confidence level, that the equipment will
perform its intended functions satisfactorily or without failure, i.e., within specified
performance limits, at a given age, for a specified length of time, function period,
or mission time, when used in the manner and for the purpose intended while
operating under the specified application and operation environments with their
associated stress levels. ”2
With all of the conditions removed, this boils down to defining reliability as the
ability of a product to perform its intended mission without failing. The definition
of reliability springs directly from the product mission, in that product failure is the
inability of the product to perform its defined mission.
2.2.1 RELIABILITY SPECIFICATIONS

In order to develop a good reliability program for a product, the product must
have good reliability specifications. These specifications should address most, if
not all, of the conditions in the reliability definition above, including mission time,
usage limitations, operating environment, etc. In many instances, this will require
a detailed description of how the product is expected to perform reliability-wise.
Use of a single metric, such as MTBF, as the sole reliability metric is inadequate.
Even worse is the specification that a product will be "no worse" than the
previous model. An ambiguous reliability specification leaves a great deal of
room for error, and this can result in a poorly-understood and unreliable product
reaching the field.
Of course, there may be situations in which an organization lacks the reliability

background or history to easily define specifications for a product's reliability. In
these instances, an analysis of existing data from previous products may be
necessary. If enough information exists to characterize the reliability performance
of a previous product, it should be a relatively simple matter to transform this
2
Kececioglu, Dimitri, Reliability Engineering Handbook, Vol. 1, Prentice-Hall, 1991.

historical product reliability characterization into specifications of the desired

reliability performance of the new product.
Financial concerns will definitely have to be taken into account when formulating
reliability specifications. Planning for warranty and production part costs is a
significant part of financial planning for the release of a new product. Based on
financial inputs such as these, a picture of the required reliability for a new
product can be established. However, financial wishful thinking should not be the
sole determinant of the reliability specifications. It can lead to problems such as
unrealistic goals, specifications that change on a regular basis to fit test results,
or test results that get "fudged" in order to conform to unrealistic expectations. It
is necessary to couple the financial goals of the product with a good
understanding of product performance in order to get a realistic specification for
product reliability. A proper balance of financial goals and realistic performance
expectations are necessary to develop a detailed and balanced reliability
specification.
2.3 UNIVERSAL FAILURE DEFINITIONS

Another important foundation for a reliability program is the development of
universally agreed-upon definitions of product failure. This may seem a bit silly,
in that it should be fairly obvious whether a product has failed or not,3 but such a
definition is quite necessary for a number of different reasons.
One of the most important reasons is that different groups within the organization
may have different definitions as to what sort of behavior actually constitutes a
failure. This is often the case when comparing the different practices of design
and manufacturing engineering groups. Identical tests performed on the same
product by these groups may produce radically different results simply because
the two groups have different definitions of product failure. In order for a reliability
program to be effective, there must be a commonly accepted definition of failure
for the entire organization. Of course, this definition may require a little flexibility
depending on the type of product, development phase, etc., but as long as
everyone is familiar with the commonly accepted definition of failure,
communications will be more effective and the reliability program will be easier to
manage.
Another benefit of having universally agreed-upon failure definitions is that it will

minimize the tendency to rationalize away failures on certain tests. This can be a
problem, particularly during product development, as engineers and managers
may tend to overlook or diminish the importance of failure modes that are
unfamiliar or not easily replicable. This tendency is only human, and a person
who has spent a great deal of time developing a product may find justification for
writing off oddball failures as a "glitch" or as failure due to some other external
error. However, this type of mentality also results in products with poorly-defined
but very real failure modes being released into the field. Having a specific failure
definition that applies to all or most types of tests will help to alleviate this
problem.
However, a degree of flexibility is called for in the definition of failure, particularly

with complex products that may have a number of distinct failure modes. For this
3
This is closely related to the concept of product mission. A good baseline definition of failure is the inability
of the product to perform its mission. From that basic definition, more detailed categorizations of failure can
be developed.

reason, it may be advisable to have a multi-tiered failure definition structure that

can accommodate the behavioral vagaries of complex equipment.
The following three-level list of failure categories is provided as an example:
• Type I – Failure: Severe operational incidents that would definitely

result in a service call, such as part failures, unrecoverable
equipment hangs, DOAs, consumables that fail/deplete before their
specified life, onset of noise, and other critical problems. These
constitute “hard-core” failure modes that would require the services
of a trained repair technician to recover.
• Type II – Intervention: Any unplanned occurrence or failure of
product mission that requires the user to manually adjust or
otherwise intervene with the product or its output. These tend to be
“nuisance failures” that can be recovered by the customer, or with
the aid of phone support. Depending on the nature of the failure
mode, groups of the Type II failures could be upgraded to Type I if
they exceed a predefined frequency of occurrence.
• Type III – Event: Events will include all other occurrences that do not
fall into either of the categories above. This might include events that
cannot directly be classified as failures, but whose frequency is of
engineering interest and would be appropriate for statistical analysis.
Examples include failures caused by test equipment malfunction or
operator error.
During testing, all of these occurrences will be logged with codes to separate the
three failure types. Other test-process-related issues such as deviations from test
plans will be logged in a separate test log. There should be a timely review of
logged occurrences to insure proper classification prior to metric calculation and
reporting. These and other procedural issues will be discussed in detail in the
next section.

3 RELIABILITY TESTING
Reliability testing is the cornerstone of a reliability engineering program. It
provides the most detailed forms of data in that the conditions under which the
data are collected can be carefully controlled and monitored. Furthermore, the
reliability tests can be designed to uncover particular suspected failure modes
and other problems. The type of reliability testing a product undergoes will
change along different points of its life cycle, but the overriding goal is to insure
that data from all or most of the tests were generated under similar enough
conditions so that an "apples-to-apples" comparison can be made of the
product's reliability characteristics at different points in the product's life. It is for
this reason that consistent and thorough reliability specifications and a standard
definition of failure are up-front requirements to implementing reliability testing.
A properly designed series of tests, particularly during the product's earlier

design stages, can generate data that would be useful in the implementation of a
reliability growth tracking program. This will provide information helpful in making
management decisions regarding scheduling, development cost projections and
so forth. This information will also be useful in planning the development cycle of
future products.
3.1 CUSTOMER USAGE PROFILING

An important requirement for designing useful reliability tests is to have a good
idea of how the product is actually going to be used in the field. The tests should
be based on a realistic expectation of the customer usage, rather than estimates
or "gut feelings" about the way the customer will use the product. Tests based on
mere speculation may result in a product that has not been rigorously tested and
consequently may run into operational difficulties due to use stress levels being
higher than anticipated. On the other hand, tests that are designed with a strong
basis of information on how the product will be used will be more realistic and
result in an optimized design that will exhibit fewer failures in the field.
Customer usage profiles can be designed to actively gather information on how

the customers are actually using an organization's product. This design can
range from a simple questionnaire to a sophisticated instrumentation within the
product that feeds back detailed information about its operation. An incentive is
often useful to get customers to sign on for a usage measurement program,
particularly if it is an intrusive process that involves the installation of data
collection equipment. Additionally, customers are often eager to participate in
these programs in the knowledge that the information that they provide will
ultimately result in a more reliable and user-friendly product.
3.2 TEST TYPES

In many cases, the type of testing that a product undergoes will change as the
product's design becomes mature and the product moves from the initial design
stages to final design release and production. Nevertheless, it is a good practice
to continue to collect internally-generated data concerning the product's reliability
performance throughout the life cycle of the product. This will strengthen the
reliability growth analysis and help provide correlation between internal test
results and field data. A brief summary of the various types of reliability tests is
presented next.

3.2.1 DEVELOPMENT TESTING

Development testing occurs during the early phases of the product's life cycle,
usually from project inception to product design release. It is vital to be able to
characterize the reliability of the product as it progresses through its initial design
stages so that the reliability specifications will be met by the time the product is
ready for release. With a multitude of design stages and changes that could
affect the product's reliability, it is necessary to closely monitor how the product's
reliability grows and changes as the product design matures. There are a
number of different test types that can be run during this phase of a product's life
cycle to provide useful reliability information:
• Component-level Testing: Although component-level testing can

continue throughout the development phase of a product, it is most
likely to occur very early in the process. This may be due to the
unavailability of parts in the early stages of the development
program. There may also be special interest in the performance of a
specific component if it has been radically redesigned, or if there is a
separate or individual reliability specification for that component. In
many cases, component-level testing is undertaken to begin
characterizing a product's reliability even though full system-level
test units are unavailable or prohibitively expensive. However,
system-level reliability characterization can be achieved through
component-level testing. This is possible if sufficient understanding
exists to characterize the interaction of the components. If this is the
case, the system-level reliability can be modeled based on the
configuration of components and the result of component reliability
testing.
• System-level Testing: Although the results of component-level tests
can be used to characterize the reliability of the entire system, the
ideal approach is to test the entire system, particularly if that is how
the reliability is specified. That is, if the technical specifications call
out a reliability goal for a specific system or configuration of
components, that entire system or configuration should be tested to
compare the actual performance with the stated goal. Although early
system-level test units may be difficult to obtain, it is advisable to
perform reliability tests at the system level as early in the
development process as possible. At the very least, comprehensive
system-level testing should be performed immediately prior to the
product's release for manufacturing, in order to verify design
reliability. During such system-level reliability testing, the units under
test should be from a homogeneous population and should be
devoted solely to the specific reliability test. The results of the
reliability test could be skewed or confounded by "piggybacking"
other tests along with it, and this practice should be avoided. A
properly conducted system-level reliability test will be able to provide
valuable engineering information above and beyond the raw
reliability data.
• Environmental and Accelerated Testing: It may be necessary in
some cases to institute a series of tests in which the system is tested
at extreme environmental conditions, or with other stress factors
accelerated above the normal levels of use. It may be that the
product would not normally fail within the time constraints of the test,
and, in order to get meaningful data within a reasonable time, the
stress factors must be accelerated. In other cases, it may be

necessary to simulate different operating environments based on

where the product will be sold or operated. Regardless of the cause,
tests like these should be designed, implemented and analyzed with
care. Depending on the nature of the accelerating stress factors, it is
easy to draw incorrect conclusions from the results of these tests. A
good understanding of the proper accelerating stresses and the
design limits of the product are necessary to be able to implement a
meaningful accelerated reliability test. For example, one would not
want to design an accelerated test that would overstress the product
and introduce failure modes that would not normally be encountered
in the field. Given that there have been a lot of incredible claims
about the capability of accelerated testing and the improbably high
acceleration factors that can supposedly be produced, care needs to
be taken when setting up this type of reliability testing program.
• Shipping Tests: Although shipping tests do not necessarily qualify
as reliability tests per se, shipping tests or simulations designed to
test the impact on the product of shipping and handling should be a
prerequisite to reliability testing. This is because the effects of
shipping will often have an impact on the reliability of the product as
experienced by the customer. As such, it may be useful to
incorporate shipping tests alongside the normal reliability testing. For
example, it may be a good idea to put the units of a final design
release reliability test through a non-destructive shipping test prior to
the actual reliability testing in order to better simulate actual use
conditions.
3.2.2 MANUFACTURING TESTING

The testing that takes place after a product design has been released for
production generally tends to measure the manufacturing process rather than the
product, under the assumption that the released product design is final and good.
However, this is not necessarily the case, as post-release design changes or
feature additions are not uncommon. It is still possible to obtain useful reliability
information from manufacturing testing without diluting any of the process-
oriented information that these tests are designed to produce.
• Functionality Testing and Burn-In: This type of testing usually falls

under the category of operation verification. In these tests, a large
proportion, if not all, of the products coming off of the assembly line
are put on a very short test in order to verify that they are functioning.
In some situations, they may be run for a predetermined "burn-in"
time in order to weed out those units that would have early infantile
failures in the field. Although it may not be possible to collect detailed
reliability information from this type of testing, what is lost in quality is
made up for in quantity. With the proper structuring, these tests can
provide a fairly good picture of early-life reliability behavior of the
product.
• Extended Post-Production Testing: This type of testing usually
gets implemented near the end or shortly after the product design is
released to production. It is useful to structure these types of tests to
be identical to the final reliability verification tests conducted at the
end of the design phase. The purpose of these tests is to assess the
effects of the production process on the reliability of the product. In
many cases, the test units that undergo reliability testing prior to the
onset of actual production are hand-built or carefully adjusted prior to

the beginning of the reliability tests. By replicating these tests with

actual production units, potential problems in the manufacturing
process can be identified before many units are shipped.
• Design/Process Change Verification: This type of testing is
similar to the extended post-production testing in that it should
closely emulate the reliability verification testing that takes place at
the end of the design phase. This type of testing should occur at
regular intervals during production, or immediately following a post-
release design change or a change in the manufacturing process.
These changes can have a potentially large effect on the reliability of
the product, and these tests should be adequate, in terms of duration
and sample size, to detect such changes.

4 FIELD DATA
While reliability testing is vital to the implementation of a reliability program, it is
not the sole source of product reliability performance data. Indeed, the
information received from the field is the "true" measure of product performance,
and is directly linked to the financial aspects of a product. In fact, a significant
proportion of field data may be more finance-related than reliability-related.
However, given the importance of the link between reliability and income, it is
important to insure that adequate reliability information can be gleaned from field
performance data. In many cases, it is not too difficult to adapt field data
collection programs to include information that is directly applicable to reliability
reporting.
Some of the most prevalent types of field data are: sales and forecasting data,
warranty data, field service data, customer support data, and returned
parts/failure analysis data, as discussed below. These discussions will tend
towards generalizations, as every organization has different methods of
monitoring the performance of its products once they are in the field. However,
the illustrations below give a good general overview of how different types of field
data may be collected and put to use for a reliability program.
4.1 THE “DISCONNECT” BETWEEN IN-HOUSE AND FIELD DATA

It should be noted at this point that there will usually be a "disconnect," or
seeming lack of correlation, between the reliability performance of the products in
the field and the results of in-house reliability testing. A typical rule of thumb is to
expect the reliability in the field to be half of what was observed in the lab. Some
of the specific causes of this disparity are discussed below, but in general the
product will usually receive harsher treatment in the field than in the lab. Units
being tested in the labs are often hand-built or carefully set up and adjusted by
engineers prior to the beginning of the test. Furthermore, the tests are performed
by trained technicians who are adept at operating the product being tested. Most
end-use customers do not have the advantage of a fine-tuned unit and training
and experience in its operation, thus leading to many more operator-induced
failures than were experienced during in-house testing. Also, final production
units are subject to manufacturing variation and transportation damage that test
units might not undergo, leading to yet more field failures that would not be
experienced in the lab. Finally, the nature of the data that goes into the
calculations will be different; in-house reliability data is usually a great deal more
detailed than the catch-as-catch-can data that characterizes a great deal of field
data.4 As can be seen, there are any number of sources for the variation
between field reliability data and in-house reliability test results. However, with
careful monitoring and analysis of both sources of data, it should be possible to
model the relationship between the two, allowing for more accurate prediction of
field performance based on reliability testing results.
4.2 SALES AND FORECASTING (SHIPPING) DATA

The sales and forecasting category of data is a sort of general-use data that is
necessary as a basis for many other analyses of field data. Essentially, this
information provides the analyst with a figure for the population of products in the
4
This problem can be ameliorated by setting up a comprehensive field data collection system. This is
discussed in more detail in the section on Field Data Collection.

field. Knowing how many units are being used at any given time period is
absolutely vital to performing any sort of reliability-oriented calculations. Having
an accurate measurement of the number of failures in the field is basically
useless if there is not a good figure for the total number of units in the field at that
time.
4.3 WARRANTY DATA

The Warranty Data category is somewhat of a catch-all category that may or may
not include the other types of field data listed below, and may not contain
adequate information to track reliability-related data. Since most warranty
systems are designed to track finances and not performance data, some types of
warranty data may have very little use for reliability purposes. However, it may be
possible to acquire adequate reliability information based on the inputs of the
warranty data, if not the actual warranty data itself. For example, a warranty
system may have ship dates and service call dates, but not actual time-to-failure
data. In this case, we must make the assumption that the failure time is
approximately equal to the difference between the ship date and service call
date, even though the product may not have actually been used during the extent
of that time before it failed. This of course is a case of "garbage in, garbage out,"
and a poorly designed warranty tracking system will yield poor or misleading data
regarding the reliability of the product. At the very least, there should be a degree
of confidence regarding the raw number of failures or warranty hits during a
particular time period. This, coupled with accurate shipping data, will allow a
crude approximation of reliability based on the number of units that failed versus
the number of units operating in the field in any given time period.
4.4 FIELD SERVICE DATA

Field Service data is a category of data that is connected with field service calls
where a repair technician manually repairs a failed product during an on-site visit.
This is a potentially powerful source of field reliability information, if a system is in
place to gather the necessary data during the service call. However, the job of
the service technician is to restore the customer's equipment to operating
condition as quickly as possible, and not necessarily to perform a detailed failure
analysis. This can lead to a number of problems. First, the service technician
may not be recording information necessary to reliability analysis, such as how
much time the product accumulated before it failed. Second, the technician may
take a "scattershot" approach to repair. That is, based on the failure symptom,
the technician will replace all of the parts whose failure may result in the failure of
that particular system. It may be that only one of the parts that were replaced had
actually failed, so it is necessary to perform a failure analysis on all of the parts to
determine which one was actually the cause of the product failure. Unfortunately,
this is not always done, and if it is, the parts that have had no problem found with
them will often be returned to field service circulation. This may lead to another
potential source of error in field service data, in that used parts with unknown
amounts of accumulated time and damage may be used as replacement parts on
subsequent service calls. This makes tracking and characterizing field reliability
very difficult. From a reliability perspective, it is always best to record necessary
failure information, avoid using the "scattershot" approach to servicing failed
equipment, and always use new units when making part replacements.

4.5 CUSTOMER SUPPORT DATA

Customer Support data is a category of data that comes from phone-in customer
support services. In many cases, it may be directly related to the field service
data in that the customer with a failed product will call to inform the organization.
In some circumstances, it may be possible to solve the customer's problem over
the phone, or adequately diagnose the cause of the problem so that a
replacement part may be sent directly to the customer without requiring a service
technician to make an on-site visit. Ideally, the customer support and field service
data would reside in the same database, but this is not always the case.
Regardless of the location, customer support data must always be screened with
care, as the information does not always reflect actual problems with the product.
Many customer support calls may concern usability issues or other instances of
the customer not being able to properly use the product. In cases such as this,
there will be a cost to the organization or warranty hit, even though there is no
real fault or failure for the product. For example, a product that is very reliable,
but has a poorly written user manual may generate a great deal of customer
support calls. This is because, even though the product is working perfectly, the
customers are having difficulty operating the product. This is a good example of
one of the sources of the "disconnect" between in-house and field reliability data.
4.6 RETURNED PARTS/FAILURE ANALYSIS DATA

As was mentioned earlier, failed parts or systems are sometimes returned for
more detailed failure analysis than can be provided by the field service
technician. Data from this area are usually more detailed regarding the cause of
failure, and are usually more useful to design or process engineers than to
reliability engineers. However, it is still an important source of information
regarding the reliability behavior of the product. This is especially true if the field
service technicians are using the “scattershot” approach to servicing the failed
product, replacing a number of parts that may or may not be defective. If this is
the case, it is necessary for all of the returned parts to be analyzed to determine
the true cause of the failure. The results of the failure analysis should be
correlated with the field service records in order to provide a complete picture of
the nature of the failure. Often this correlation does not occur, or the returned
parts are not analyzed in a timely fashion. Even if the analysis is performed
correctly, there tend to be a significant proportion of returned parts with which no
problem can be found. This is another example of a potential cause of the
disparity between lab and field reliability data. However, even if the failure
analysis group is unable to assign a cause to the failure, a failure has taken
place, and the organization has taken a warranty hit. In the field, the performance
the customer experiences is the final arbiter of the reliability of the product.

5 DATA COLLECTION
Data collection is the framework for a good reliability engineering program. It is
necessary to have an accurate and comprehensive system of recording data
relating to a product's reliability performance in order to be able to produce
meaningful reliability reports. Although the nature of the data collection system
will differ based on the type of data being collected, there must be a certain
number of common elements in the type of data being collected and the way the
information is recorded. This is necessary in order to provide the continuity
necessary for developing an accurate assessment of the "cradle-to-grave"
reliability of a product.
Whenever possible, computers should be employed in the data collection and

recording process. Of course, the method of data collection will vary with the
product under consideration, but given the decreasing cost and increasing power
of computer systems, it should not be very difficult to set up a computerized data
collection system. In some cases, it is even possible to automate much of the
data collection process, thus further decreasing the potential for data recording
errors. The concepts of in-house data collection, field data collection, and
ReliaSoft's Dashboard are presented in more detail below.
5.1 IN-HOUSE TEST DATA COLLECTION

In a previous section, the different types of in-house reliability testing were
discussed. One of the most important aspects of setting up an in-house reliability
testing program lies in having certain common elements that extend across all of
the different types of tests. This can be aided greatly by a degree of uniformity in
the data collection process. By having a core group of data types that are
collected from every test that takes place, it is easier to perform similar analyses
across a variety of different test types. This lends a great deal of continuity in test
analysis and reporting that will benefit the reliability program and the entire
organization.
As mentioned earlier, it is highly beneficial to automate the data collection and

recording process wherever possible. The method of data collection will differ
from product to product, depending on whether it is possible to use a computer
interface to operate the product during the test and automatically record the test
results. Regardless of the method employed in running the test, it is always
possible and advisable to use a database system to keep track of the test
results. The use of relational databases makes it fairly easy to manipulate large
quantities of data, and greatly aids in the reporting process. Properly managed, it
is even possible to automate some if not all of the data reporting process using
database information. Of course, human oversight is always a necessity when
dealing with data analysis of any type, but proper use of database structuring
and manipulation can make the reliability engineer's job much easier.
In setting up a database system for collecting data from in-house reliability

testing, there are a minimum number of data types that need to be included in
the database structure. For the purposes of in-house reliability data collection, it
is recommended to have at least three related databases: a test log, a failure log,
and a service log. Detailed descriptions of these databases and the information
they should contain appear below.

5.1.1 TEST LOG

The test log contains detailed information on the tests being run on the products.
The structure of the database will vary depending on the testing procedures and
depending on the type of products for which data are being captured. If the
product requires a test in which the test units are essentially just turned on and
left to run until they fail or the time of the test expires, the test log will be fairly
simple. However, if the product requires a variety of different inputs in order to be
properly exercised during testing, the test log should be detailed enough to
record all of the pertinent information. A suggested list of fields for the test log
include:
• Transaction number: a unique identification code for the test log

entry.
• Test start date: the date the test starts.
• Test start time: the time the test starts.
• Test name: the name or identifier for the test being run.
• Test stage or step: if the test is run in a series of stages or steps
with different inputs, this field should provide a description or count
of which segment of the test is being run, e.g. “Step 2,” “High
Temperature,” etc.
• Test inputs: this describes the test inputs at each stage or step of
the test. Depending on the nature of the product and the testing, it
may be necessary to create a separate log for the specific steps and
inputs of the test in order to keep the test log from being too cluttered
with specific step/input information.
• Operator comments: specific comments regarding the test that
may be useful when performing subsequent analyses.
• Deviations: descriptions of deviations from the original test plan.
5.1.2 FAILURE LOG

The failure log is where the majority of the information that is important to the
generation of reliability results will reside. Care should be taken in the
construction of this database so that all of the pertinent failure information will be
collected every time a failure occurs. At the same time, it should not have so
many fields as to be unwieldy when conducting a reliability analysis. This might
slow down the overall testing process if a large amount of minutely detailed
information needs to be recorded. Developing a method of automating the data
collection process will alleviate this problem, but that is not always possible. A
suggested list of fields for the failure log include:
• Transaction number: a unique identification code for the failure log

entry.
• Test log cross-reference: the transaction number for the test log
entry which corresponds to the test on which the failure occurred.
• Service log cross-reference: the transaction number for the most
recent service log entry.
• Failure date: the date when the failure occurred.
• Failure time: the time when the failure occurred.

• Failure type: this describes the failure type encountered,

particularly if a multi-tiered system of failure classification is being
used.
• Test stage: the stage or step of the test when the failure occurred.
This can be cross-referenced to the appropriate test or input log.
• Symptom code: this is the symptom noticed by the operator when
the failure occurred. The type of symptom code can be cross-linked
to a preliminary failure code.
• Failure code: this is a code that describes the actual mode of
failure. A preliminary failure code can be generated based on the
symptom code, but a failure analysis engineer should make the final
disposition of the failure code.
• Failed part ID: this describes the part or parts that caused the
failure. If possible, the failed part serial number should be included
as a separate field.
• Resolution: this field describes what action was taken to restore
the failed unit to operational status. This field should be cross-linked
to the service log.
• Comments: specific comments regarding the failure that may be
useful when performing subsequent analyses.
5.1.3 SERVICE LOG

The purpose of the service log is to track and record any service actions or
modifications performed on test units. It is important to keep a concise record of
any service actions performed on test units because even a relatively small
modification or repair can potentially have a large effect on the performance of
the test units. By requiring service technicians and engineers to use a service log
whenever they work on a test unit, the amount of unofficial "tinkering" with a
system will be minimized, thus reducing unexplained changes in test unit
performance. A service log entry should be made whenever a test unit is
installed or upgraded. This allows for tracking design level or version number
changes across the tests. A suggested list of fields for the service log include:
• Transaction number: a unique identification code for the service

log entry.
• Test log cross-reference: the transaction number for the test log
entry which corresponds to the test during which the service was
performed.
• Service date: the date on which the service was performed.
• Service time: the time at which the service was performed.
• Current version identifier: identifies the revision or design level of
the test unit before the service is performed.
• New version identifier: identifies the revision or design level of the
test unit after the service is performed. This will be the same as the
current version identifier unless the service performed upgrades the
test unit to the next level.
• Service type: describes the service performed.
• Part modified/replaced: a description/serial number of the part
modified or replaced during the service action.

• Comments: specific comments regarding the service action that

may be useful when performing subsequent analyses.
5.2 FIELD DATA COLLECTION

Depending on the circumstances, collection of field data for reliability analyses
can be either a simple matter or major headache. Even if there is not a formal
field data collection system in place, odds are that much of the necessary
general information is being collected already in order to track warranty costs,
financial information, etc.. The potential drawback is that the data collection
system may not be set up to collect all of the types of data necessary to perform
a thorough reliability analysis. As mentioned earlier, many field data collection
methodologies focus on aspects of the field performance other than reliability.
Usually, it is a small matter to modify data collection processes to gather the
necessary reliability information.
For example, in one instance the field repair personnel were only collecting
information specific to the failure of the system and what they did to correct the
fault. No information was being collected on the time accumulated on the
systems at the time of failure. Fortunately, it was a simple matter to have the
service personnel access the usage information, which was stored on a
computer chip in the system. This information was then included with the rest of
the data collected by the service technician, which allowed for a much greater
resolution in the failure times used in the calculation of field reliability. Previously,
the failure time was calculated by subtracting the failure date from the date the
product was shipped. This could cause problems in that the product could remain
unused for months after it was shipped. By adding the relatively small step of
requiring the service technicians to record the accumulated use time at failure, a
much more accurate model of the field reliability of this unit could be made.
Another difficulty in using field data to perform reliability analyses is that the data
may reside in different places, and in very different forms. The field service data,
customer support data, and failure analysis data may be in different databases,
each of which may be tailored to the specific needs of the group recording the
data. The challenge in this case is in developing a method of gathering all of the
pertinent data from the various sources and databases and pulling it into one
central location where it can easily be processed and analyzed. These functions
can also be performed automatically, using ReliaSoft's Dashboard system.
5.2.1 RELIASOFT’S DASHBOARD

ReliaSoft's Dashboard system is a tool for the automation of product quality
tracking and warranty processes that pulls in data from a variety of sources and
presents the analyzed results in a central location. It is designed around a central
database that is used to capture, analyze and present product field reliability,
quality and warranty data. The system can be used to capture product quality
data for product failures reported via customer returns, the customer call center,
field repairs, and other warranty channels. The ReliaSoft Dashboard is a Web-
based (or client-server) reporting mechanism that allows users to view product
quality and reliability reports and analyses. As needed, customized data entry
tools, data load applications, and data transfers from existing systems can be
used to capture product data. Following is a description of a Dashboard system
currently in use to capture and track field reliability information.
The system is designed around a master database that captures product quality
and reliability data into one centralized location. Data are incorporated into the

master database through a variety of techniques designed to work with the

information infrastructure already in place at the organization. In this case, Sales,
Manufacturing, Warranty Cost, Call Center and Repair Center data are extracted
from the databases in which they currently reside on a regular basis and
"pumped" into the master database using a specially designed data import utility
to validate and load the data. (Other methods to transfer data from existing
databases are available and their implementation depends on the existing
information architecture and the information technology policies of the individual
organization.)
Detailed data on product returns (processed through regional distribution

organizations) are captured via a customized Web-based data entry interface
that serves to validate and load the data into the master database. Some of the
products returned by customers are routed through a product disposition center
that has been designed to coordinate more detailed analysis of returned products
based on statistical sampling techniques. A customized application in use at this
location allows the organization to coordinate the disposition and sampling. More
extensive failure analyses are performed on selected products and data obtained
during that process is incorporated into the system via a Web-based data entry
interface as well. A graphical representation of an example of this process is
shown in Figure 2.
Once the pertinent information has been loaded to a central storage area, it can
easily be processed and presented in a format that will be meaningful to the
users of the information. The Dashboard system presents an at-a-glance
overview of a variety of different analyses based on the data that have been
extracted. It is then possible to "drill down" to more detailed levels of information
from the top-level view.
Another advantage to using the Dashboard (or any other tool or methodology) to
pull together disparate sources of field information is that the planning involved in
bringing these previously separate sources of information together usually results
in a beneficial synergy. By bringing together key players and acting as a catalyst
to discussion and discovery, the planning process often helps an organization to
gain a new understanding of its current processes and to identify untapped
existing resources. Many organizations that have undergone this process have
found that the fresh perspective and ability to spark communication provided by
such consultations are invaluable to their organizations.

MFG
SALES Master Database DASHBOARD

& Reports &
Forecast Analyses
Warranty
Costs
Corporate Repair
Center(s)
Detailed
Failure Analysis
by Product Engineers
Corporate
Technical
Hotline(s)
3rd-Party
Field Repairs
Product
Disposition
Center
Electronic data transmission

Custom applications for data
Product disposition entry and access.
Commercial Other data on paper forms, Direct data extraction from

Distributors electronic communication, etc. existing information system.
Legend
Figure 2 - Example of Linking Existing Data Sources and Customized Data Sources for a
Dashboard System

6 DATA ANALYSIS AND REPORTING

The manner in which reliability data is analyzed and reported will largely have to
be tailored to the specific circumstance or organization. However, it is possible to
break down the general methods of analysis/reporting into two categories:
parametric analyses and non-parametric analyses. Overall, it will be necessary to
tailor the analysis and reporting methods by the type of data as well as to the
intended audience. Managers will generally be more interested in actual data
and non-parametric analysis results, while engineers will be more concerned
with parametric analysis. Of course this is a rather broad generalization and if the
proper training has instilled the organization with an appreciation of the
importance of reliability engineering, there should be an interest in all types of
reliability reports, at all levels of the organization. Nevertheless, managers are
usually more interested in the "big picture" information that non-parametric
analyses generally tend to provide, while not being particularly interested in the
level of technical detail that parametric analyses provide. On the other hand,
engineers and technicians are usually more concerned with the close-up details
and technical information that parametric analyses provide. Both of these types
of data analysis have a great deal of importance to any given organization, and it
is merely necessary to apply the different types in the proper places.
6.1 NON-PARAMETRIC ANALYSIS

Data conducive to non-parametric analysis is information that has not or cannot
be rigorously "processed" or analyzed. Usually, it is simply straight reporting of
information, or if it has been manipulated, it is usually by simple mathematics,
with no complex statistical analysis. In this respect, many types of field data lend
themselves to the non-parametric type of analysis and reporting. In general, this
type of information will be of most interest to managers as it usually requires no
special technical know-how to interpret. Another reason it is of particular interest
to managers is that most financial data falls into this category. Despite its relative
simplicity, the importance of non-parametric data analysis should not be
underestimated. Most of the important decisions that are made concerning the
business are based on non-parametric analysis of financial data.
As mentioned in the previous section, ReliaSoft's Dashboard system is a

powerful tool for collecting and reporting data. It especially lends itself to non-
parametric data analysis and reporting, as it can be quickly processed and
manipulated in accordance with the user's wishes.
6.1.1 NON-PARAMETRIC RELIABILITY ANALYSIS

Although many of the non-parametric analyses that can be performed based on
field data are very useful for providing a picture of how the products are behaving
in the field, not all of this information can be considered "hard-core" reliability
data. As was mentioned earlier, many such data types and analyses are just
straight reporting of the facts. However, it is possible to develop standard
reliability metrics such as product reliability and failure rates from the non-
parametric analysis of field data. A common example of this is the "diagonal
table" type of analysis that combines shipping and field failure data in order to
produce empirical measures of defect rates.
Table 1 gives an example of a "diagonal table" of product shipping and failure

data by shipment week. The top row, highlighted in blue and yellow, shows the
number of units of product that were shipped in a given week, labeled from 9901
to 9920. The data highlighted in blue and gray represents the number of units

that were reported failed or had warranty hits in the subsequent weeks after
being shipped. This information can be used to calculate a simple percent
defective for each shipment week. Note that one must make certain to use a
weighting factor to account for the amount of time a particular week's worth of
units have spent in the field. Also, care should be taken to account for the delay
between shipping and actual installation, which can be a substantial time period
for some products. The time period for the average delay (in this example two
weeks, the data of which appears in the gray diagonal in the table) should be
removed from the data being analyzed. Otherwise, a false appearance of a
decreasing defect rate appears in the final results of the analysis. Figure 4 shows
the results of the non-parametric defect rate calculation, unadjusted and adjusted
for the two-week average delay between shipping and installation.
Table 1 - "Diagonal Table" of Field Data for Non-Parametric Analysis

Product Defective Rate by Ship Week

With Data Unadjusted and Adjusted for Install Delay
9%
8%
Percent Defective
7%
6%
5%
4%
3%
2%
1%
0%
9901
9902
9903
9904
9905
9906
9907
9908
9909
9910
9911
9912
9913
9914
9915
9916
9917
9918
9919
9920
Ship Week
Unadjusted Adjusted
Figure 3 - Percent Defective Results from Data in Table 1, Unadjusted and Adjusted for
Installation Delay
6.2 PARAMETRIC ANALYSIS

Data that lends itself to parametric statistical analysis can produce very detailed
information about the behavior of the product based on the process utilized to
gather the data. This is the "hard-core" reliability data with all the associated
charts, graphs and projections that can be used to predict the behavior of the
products in the field.
The origin of this type of data is usually in-house, from reliability testing done in
laboratories set up for that specific purpose. For that reason, a great deal more
detail will be associated with these data sets than with those that are collected
from the field. Unfortunately, in dealing with field data, it is often a matter of
taking what you can get, without being able to have much impact on the quality
of the data. Of course, setting up a good program for the collection of field data
will raise the quality of the field data collected, but generally it will not be nearly
as concise or detailed as the data collected in-house.
The exception to this generalization is field data that contains detailed time-of-
use information. For example, automotive repairs that have odometer
information, aircraft repairs that have associated flight hours or printer repairs
that have a related print count can lend themselves to parametric analysis.
Caution should be exercised when this type of analysis is performed, however, to
make sure that the data are consistent and complete enough to perform a
meaningful parametric analysis.
Although it is possible to automate parametric analysis and reporting, care

should be taken in automatic processing. Caution is required because of the
level of detail inherent in this type of data and the potential "disconnect" between
field data and in-house testing data (described in Section 4.1). Presentations of
these two types of data should be carefully segregated in order to avoid

unnecessary confusion among the end users of the data reports. It is not unusual
for end-users who are not familiar with statistical analyses to become confused
and indignant when presented with seemingly contradictory data on a particular
product. The tendency in cases such as these is to accuse one or both sources
of data (field or in-house) of being inaccurate. This is, of course, not necessarily
true. As was discussed earlier, there will usually tend to be a disparity between
field data and in-house reliability data.
Another reason for the segregation of the field data and the in-house data is the
need for human oversight when performing the calculations. Field data sets tend
to undergo relatively simple mathematical processing which can be safely
automated without having to worry about whether the analysis type is appropriate
for the data being analyzed. However, this can be a concern for in-house data
sets that are undergoing a more complicated statistical analysis. This is not to
say that parametric analysis should not be in any way automated. However, a
degree of human oversight should be included in the process to insure that the
data sets are being analyzed in an appropriate manner. Furthermore, the data
should be cross-referenced against the Test Log and Service Log to make sure
that irrelevant or "outlier" information is not being included in the data.
6.2.1 EXAMPLES OF REPORTING FOR PARAMETRIC DATA ANALYSIS

Following are some examples of the information that can be generated using
parametric data analysis. While this is by no means complete, it serves as a
starting point for the information that can be obtained with the proper collection of
data and parametric analysis.
6.2.1.1 PROBABILITY PLOT

Probability plotting was originally a method of graphically
Generated by: ReliaSoft's W eibull++ 5.0 - www.W eibull.com - 888-886-0410
Probability Plot
99.00
Weibull
Data 1
estimating distribution parameter values. With the use of
90.00
P=2, A=RRX-S
F=19 | S=0
computers that can precisely calculate parametric values,
the probability plot now serves as a graphical method of
assessing the goodness of fit of the data to a chosen
50.00
)
(t
F
y
,
til
i
distribution. Probability plots have nonlinear scales that will
b
ail
er
n 10.00
essentially linearize the distribution function, and allow for
U
assessment of whether the data set is a good fit for that
5.00
particular distribution based on how close the data points
USER
RELIASOFT come to following the straight line. The y-axis usually shows
9/29/99
1.00
10.00 100.00 1000.00
10:21:40 AM
10000.00
the unreliability or probability of failure, while the x-axis
Time, (t)
shows the time or ages of the units. Specific characteristics
β=0.88, η=1220.90, ρ =1.00
of the probability plot will change based on the type of
distribution..

6.2.1.2 RELIABILITY FUNCTION

Generated by: ReliaSoft's Weibull++ 5.0 - www.Weibull.com - 888-886-0410 The reliability function gives the continuous probability
1.00
Reliability vs Time Plot of a successful mission versus the time of the mission.
Weibull
Data 2 This is similar to the probability plot in that it shows the
P=2, A=RRX-S
F=25 | S=0
performance of the product versus the time. However, it
0.80
does not have nonlinear scales on the axes and the y-
)
(t
axis gives the reliability instead of the unreliability.
F- 0.60
1
=)
(t
R
,
y
til
i
b 0.40
ali
e
R
0.20
USER
RELIASOFT
9/29/99
0 10:29:50 AM
0 480.00 960.00 1440.00 1920.00 2400.00
Time, (t)
β=2.17, η=1029.49, ρ =0.99
6.2.1.3 PROBABILITY DENSITY FUNCTION

Generated by: ReliaSoft's W eibull++ 5.0 - www.W eibull.com - 888-886-0410
The probability density function (pdf) represents the
Probability Density Function relative frequency of failures with respect to time. It
1.00E-3
Weibull
Data 2
basically gives a description of how the entire
P=2, A=RRX-S population from which the data is drawn is spread out
F=25 | S=0
8.00E-4
over time or usage. The probability density function is
most commonly associated with the "bell curve," which
6.00E-4
is the shape of the pdf of the normal or Gaussian
)
(t
f
distribution.
4.00E-4
2.00E-4
USER
RELIASOFT
9/29/99
0 10:39:28 AM
0 480.00 960.00 1440.00 1920.00 2400.00
Time, (t)
β=2.17, η=1029.49, ρ =0.99
6.2.1.4 FAILURE RATE FUNCTION

Generated by: ReliaSoft's W eibull++ 5.0 - www.W eibull.com - 888-886-0410 The failure rate function indicates how the number of
Failure Rate vs Time Plot
1.00E-3
failures per unit time of the product changes with time.
Weibull
Data 1 This provides a measure of the instantaneous
9.20E-4
P=2, A=RRX-S
F=19 | S=0 probability of product failure changes as usage time is
accumulated. The failure rate plot is associated with
R
)
(t
8.40E-4
the "bathtub curve," which is an amalgamation of
)/
(t
f
,
different failure rate curves which illustrates the
et
a
R
er
different ways in which products exhibit failure
7.60E-4
uli
a
F
characteristics over the course of their lifetimes.
6.80E-4
USER
RELIASOFT
9/29/99
6.00E-4 10:47:52 AM
0 800.00 1600.00 2400.00 3200.00 4000.00
Time, (t)
β=0.88, η=1220.90, ρ =1.00

6.2.1.5 LIFE VS. STRESS

The Life vs. Stress plot is a product of accelerated life
Generated by: ReliaSoft's ALTA - www.ReliaSoft.com - 888-886-0410
Life vs Stress
1.00E+5
T-H/Weib
testing or reliability testing that is performed at
Data 2
Eta
different stress levels. This indicates how the life
Mean Lif e
99 performance of the product changes at different
1
288 / 20 stress levels. The gray shaded areas are actually pdf
F=40 | S=0
10000.00
plots for the product at different stress levels. Note
that it is difficult to make a complete graphical
ef
L
i
comparison of the pdf plots due to the logarithmic
1000.00
scale of the y-axis.
user
RELIASOFT
9/29/99
11:19:59 AM
100.00
260.00 272.00 284.00 296.00 308.00 320.00
Stress
Beta=1.8261, A=4.4736, b=20.3425, Phi=2129.1813
6.2.1.6 LIKELIHOOD FUNCTION

The likelihood function is a more esoteric function of
the data, but it is directly related to how the
parameters are calculated. The likelihood function
relates the data points to the values for the
parameters of the distribution. The maximization of
this function determines the best values for the
distribution's parameters.

6.2.1.7 RELIABILITY IMPORTANCE

Reliability importance is a measure of the
Static Reliability Importance relative weight of components in a system, with
0.20 0.198257 respect to the system's reliability value. The
higher the reliability importance a particular
component has, the larger the effect that
0.15
component has on the system's reliability. This
0.130634
measure is useful in helping to optimize system
reliability performance, as it helps identify which
components will have the greatest effect on the
0.10
overall system reliability.
0.05
0.037072
0.032412 0.031455
0.00
Part A Part B Part C Part D Part E
Time = 300
User's Name
Company
10/04/1999
11:37 AM
6.2.1.8 RELIABILITY GROWTH

Reliability growth is an important component of
a reliability engineering program. It essentially
models the change in a product's reliability
over time and allows for projections on the
change in reliability in the future based on past
performance. It is useful in tracking
performance during development and aids in
the allocation of resources. There are a
number of different reliability growth models
available that are suitable to a variety of data
types. The above chart is a graphical
representation of the logistic reliability growth
model.

7 BRINGING IT ALL TOGETHER

The preceding sections briefly outline some of the basic building blocks of a solid
reliability engineering program. These steps are all very helpful in constructing a
program that will efficiently gather information, transmit, store, analyze and report
on the reliability of an organization's products. The process will not be the same
for everyone, of course. The construction or enhancement of a reliability program
will by necessity be specially adapted according to the specific needs and
structure of the organization. As is the case with many other situations, "form
follows function," and the form of the reliability program will follow the function of
the organization that is implementing it.
However, it is necessary to make sure that the information that is generated by

the reliability program is fed back throughout the organization so that the
maximum benefits of the program can be achieved. Instituting a reliability
program merely for the sake of having a reliability program will ultimately be of no
benefit to anyone. If the reliability program is not feeding back useful information
to all of the areas of the organization that need it, it will eventually atrophy and
become just a little-utilized enclave of the larger organization. Of course, it is
unlikely that an organization that has gone to the trouble of implementing a high-
efficiency reliability program will allow such a program to wither and die, but it is
important to make sure that the reliability program's benefits reach all the areas
that it can.
In the course of this outline, we have discussed some of the more immediate
benefits of having a good reliability program in place. Examples include feeding
information back to manufacturing organizations to aid in maximizing the
efficiency of the manufacturing process and performing system-level reliability
analyses that can benefit the early stages of a development program. There are
still other methods of putting reliability information to use in order to aid the
organization beyond the obvious uses, such as decreased warranty costs.
7.1 CONNECTING FIELD AND LAB DATA

One of the most important activities that can be undertaken once a
comprehensive reliability program is in place is to be able to model the transition
between reliability data generated as a result of in-house testing and reliability
data resulting from the performance of products in the field. The causes for this
difference in results as well as the differences in the formats of the data have
been discussed elsewhere in this document. Nevertheless, the ability to bridge
the difference between these two information sources lies within the grasp of an
organization that has a good reliability program and an adequate amount of data.
Although it requires a good deal of data manipulation and mathematical analysis,
a model can be developed that will allow for the mapping of in-house reliability
data to make accurate predictions of field performance. Obviously, this is a
powerful tool that would have an important role in projecting warranty costs for
new products and the planning of future programs.
7.2 RELIABILITY GROWTH

One use of the information that a reliability program provides is the
implementation of a reliability growth study. There are numerous reliability growth
models that can be used with a variety of types of input data. The diversity of
reliability growth models and acceptable input makes this type of modeling very
flexible and it can be applied across a number of different functional areas in an

organization. For example, the detailed data generated during the development
phase of a product can be used with a parametric growth model in order to judge
whether the project will meet its reliability goal within the allotted time. Based on
the growth model results, more efficient allocation of resources on the project
could be implemented based on the expected performance of the product.
Similarly, a less complicated non-parametric growth model could be used to
assess the change of field reliability as a result of design or manufacturing
process changes once the product has been released. On a larger scale, the
reliability growth of specific product lines can be modeled over the course of
several generations of products in order to estimate the reliability and associated
warranty costs of future product lines or projects that have yet to be
implemented.
7.3 OPTIMUM DESIGN LEVEL DETERMINATION

With a good grasp of the reliability of components and systems, it is possible to
devise specifications and designs that result in the optimum level of reliability
performance. Designing a product with inexpensive and unreliable parts will
result in a product with low initial costs, but high support and warranty costs. On
the other hand, over-designing a product with costly, highly reliable parts will
result in a final product with low support and warranty costs, but that is
prohibitively expensive. Application of information from a reliability engineering
program can result in a design that balances out both of these factors, resulting
in a design reliability that minimizes the overall cost of the product. Figure 5 gives
a graphical representation of this concept.
Total Cost
Cost
Optimum Reliability
Initial Cost Post-Implementation Cost
Reliability
Figure 5 - Balancing Initial and Post-Production Costs to Determine Optimum Reliability

7.4 COMPETITIVE ASSESSMENT

The principles and practices of reliability engineering that are applied to an
organization's products in the normal course of development and production can
also be applied to the products of the competition. In setting up a competitive
reliability assessment program, a population of a competitor's products is tested
in the same manner as those in-house products in development and production.
This can provide valuable information as to the relative strengths and
weaknesses of the competitors' products. In cases where the reliability or
performance of a competitor's product is superior to those being produced by an
organization, the competitor's product can be "reverse engineered" in order to
gain insight on how the organization's product can be improved. By
understanding the performance of the entire gamut of competitive products, an
organization can go a long way towards becoming the best in the field.
7.5 MARKETING AND ADVERTISING

In the competitive business world, any edge in helping to find or increase the
number of paying customers can result in sizable financial benefits. Given two
competing products that are equal in all other respects, the edge belongs to the
product that is more reliable. As products become more sophisticated, so do the
customers, to the point where the reliability of a product is one of the main
considerations a savvy customer takes into account before making a purchase.
As a result, more and more advertising includes a reliability slant as part of the
sales pitch. From computers to sewing machines, the reliability of the product is
increasingly being used to market and sell a variety of products. Some
advertisements are now including what were once considered "esoteric"
reliability concepts such as MTBF (Mean Time Before Failure) values. With a
solid reliability program in place, the information can be used to help sell the
product as well as to develop it. This is particularly true if there is data from a
competitive assessment program that allows the sales and marketing groups to
demonstrate that their product is not only highly reliable but also much more
reliable than those of the competition.


Reliability Blueprint PDF

Uploaded by

Copyright:

Available Formats

Reliability Blueprint PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reliability Blueprint PDF

Uploaded by

Copyright:

Available Formats

ReliaSoft Corporation R&D Reports Blueprint for a Comprehensive Reliability Program

ReliaSoft Corporation R & D Reports

Last Revised on: February 3, 2003

ReliaSoft Corporation Page 1

ReliaSoft Corporation Page 2

ReliaSoft Corporation Page 1

2.1 FOSTERING A CULTURE OF RELIABILITY

ReliaSoft Corporation Page 2

implementation of reliability concepts, it will be a great deal easier to go about

2.2 PRODUCT MISSION

2.2.1 RELIABILITY SPECIFICATIONS

Of course, there may be situations in which an organization lacks the reliability

ReliaSoft Corporation Page 3

historical product reliability characterization into specifications of the desired

2.3 UNIVERSAL FAILURE DEFINITIONS

Another benefit of having universally agreed-upon failure definitions is that it will

However, a degree of flexibility is called for in the definition of failure, particularly

ReliaSoft Corporation Page 4

reason, it may be advisable to have a multi-tiered failure definition structure that

The following three-level list of failure categories is provided as an example:

• Type I – Failure: Severe operational incidents that would definitely

ReliaSoft Corporation Page 5

A properly designed series of tests, particularly during the product's earlier

3.1 CUSTOMER USAGE PROFILING

Customer usage profiles can be designed to actively gather information on how

3.2 TEST TYPES

ReliaSoft Corporation Page 6

3.2.1 DEVELOPMENT TESTING

• Component-level Testing: Although component-level testing can

ReliaSoft Corporation Page 7

necessary to simulate different operating environments based on

3.2.2 MANUFACTURING TESTING

• Functionality Testing and Burn-In: This type of testing usually falls

ReliaSoft Corporation Page 8

the beginning of the reliability tests. By replicating these tests with

ReliaSoft Corporation Page 9

4.1 THE “DISCONNECT” BETWEEN IN-HOUSE AND FIELD DATA

4.2 SALES AND FORECASTING (SHIPPING) DATA

ReliaSoft Corporation Page 10

4.3 WARRANTY DATA

4.4 FIELD SERVICE DATA

ReliaSoft Corporation Page 11

4.5 CUSTOMER SUPPORT DATA

4.6 RETURNED PARTS/FAILURE ANALYSIS DATA

ReliaSoft Corporation Page 12

Whenever possible, computers should be employed in the data collection and

5.1 IN-HOUSE TEST DATA COLLECTION

As mentioned earlier, it is highly beneficial to automate the data collection and

In setting up a database system for collecting data from in-house reliability

ReliaSoft Corporation Page 13

5.1.1 TEST LOG

• Transaction number: a unique identification code for the test log

5.1.2 FAILURE LOG

• Transaction number: a unique identification code for the failure log

ReliaSoft Corporation Page 14

• Failure type: this describes the failure type encountered,

5.1.3 SERVICE LOG

• Transaction number: a unique identification code for the service

ReliaSoft Corporation Page 15