Reliability Notes

8.
Assessing Product Reliability
8. Assessing Product Reliability

This chapter describes the terms, models and techniques used to evaluate and predict
product reliability.
1. Introduction 2. Assumptions/Prerequisites
1. Why important 1. Choosing appropriate life
2. Basic terms and models distribution
3. Common difficulties 2. Plotting reliability data
4. Modeling "physical acceleration" 3. Testing assumptions
5. Common acceleration models 4. Choosing a physical acceleration
model
6. Basic non-repairable lifetime
distributions 5. Models and assumptions for
Bayesian methods
7. Basic models for repairable systems
8. Evaluate reliability "bottom-up"
9. Modeling reliability growth
10. Bayesian methodology
3. Reliability Data Collection 4. Reliability Data Analysis

1. Planning reliability assessment tests 1. Estimating parameters from
censored data
2. Fitting an acceleration model
3. Projecting reliability at use
conditions
4. Comparing reliability between two
or more populations
5. Fitting system repair rate models
6. Estimating reliability using a
Bayesian gamma prior
Open Rubric
http://www.itl.nist.gov/div898/handbook/apr/apr.htm (1 of 2) [5/22/2001 1:27:36 PM]
Click here for a detailed table of contents

References for Chapter 8
http://www.itl.nist.gov/div898/handbook/apr/apr.htm (2 of 2) [5/22/2001 1:27:36 PM]

8. Assessing Product Reliability - Detailed

Table of Contents [8.]
1. Introduction [8.1.]
1. Why is the assessment and control of product reliability important? [8.1.1.]
1. Quality versus reliability [8.1.1.1.]
2. Competitive driving factors [8.1.1.2.]
3. Safety and health considerations [8.1.1.3.]
2. What are the basic terms and models used for reliability
evaluation? [8.1.2.]
1. Repairable systems and non-repairable populations - lifetime
distribution models [8.1.2.1.]
2. Reliability or survival function [8.1.2.2.]
3. Failure (or hazard) rate [8.1.2.3.]
4. "Bathtub" curve [8.1.2.4.]
5. Repair rate or ROCOF [8.1.2.5.]
3. What are some common difficulties frequently found with reliability data
and how are they overcome? [8.1.3.]
1. Censoring [8.1.3.1.]
2. Lack of failures [8.1.3.2.]
4. What is "physical acceleration" and how do we model it? [8.1.4.]
5. What are some common acceleration models? [8.1.5.]
1. Arrhenius [8.1.5.1.]
2. Eyring [8.1.5.2.]
3. Other models [8.1.5.3.]
6. What are the basic lifetime distribution models used for non-repairable
populations? [8.1.6.]
http://www.itl.nist.gov/div898/handbook/apr/apr_d.htm (1 of 4) [5/22/2001 1:27:25 PM]

1. Exponential [8.1.6.1.]
2. Weibull [8.1.6.2.]
3. Extreme value distributions [8.1.6.3.]
4. Lognormal [8.1.6.4.]
5. Gamma [8.1.6.5.]
6. Fatigue life (Birnbaum-Saunders) [8.1.6.6.]
7. Proportional hazards model [8.1.6.7.]
7. What are some basic repair rate models used for repairable
systems? [8.1.7.]
1. Homogeneous Poisson Process (HPP) [8.1.7.1.]
2. Non-Homogeneous Poisson Process (NHPP) - power law [8.1.7.2.]
3. Exponential law [8.1.7.3.]
8. How can you evaluate reliability from the "bottom - up" (component failure
mode to system failure rates)? [8.1.8.]
1. Competing risk model [8.1.8.1.]
2. Series model [8.1.8.2.]
3. Parallel or redundant model [8.1.8.3.]
4. R out of N model [8.1.8.4.]
5. Standby model [8.1.8.5.]
6. Complex systems [8.1.8.6.]
9. How can you model reliability growth? [8.1.9.]
1. NHPP power law [8.1.9.1.]
2. Duane plots [8.1.9.2.]
3. NHPP exponential law [8.1.9.3.]
10. How can Bayesian methodology be used for reliability
evaluation? [8.1.10.]
2. Assumptions/Prerequisites [8.2.]
1. How do you choose an appropriate life distribution model? [8.2.1.]
1. Based on failure mode [8.2.1.1.]
2. Extreme value argument [8.2.1.2.]
3. Multiplicative degradation Argument [8.2.1.3.]
4. Fatigue life (Birnbaum-Saunders) model [8.2.1.4.]

5. Empirical model fitting - distribution free (Kaplan-Meier)

approach [8.2.1.5.]
2. How do you plot reliability data? [8.2.2.]
1. Probability plotting [8.2.2.1.]
2. Hazard and cum hazard plotting [8.2.2.2.]
3. Trend and growth plotting (Duane plots) [8.2.2.3.]
3. How can you test reliability model assumptions? [8.2.3.]
1. Visual tests [8.2.3.1.]
2. Goodness of fit tests [8.2.3.2.]
3. Likelihood ratio tests [8.2.3.3.]
4. Trend tests [8.2.3.4.]
4. How do you choose an appropriate physical acceleration model? [8.2.4.]
5. What models and assumptions are typically made when Bayesian methods
are used for reliability evaluation? [8.2.5.]
3. Reliability Data Collection [8.3.]

1. How do you plan a reliability assessment test? [8.3.1.]
1. Exponential life distribution (or HPP model) tests [8.3.1.1.]
2. Lognormal or Weibull tests [8.3.1.2.]
3. Reliability growth (Duane model) [8.3.1.3.]
4. Accelerated life test [8.3.1.4.]
5. Bayesian gamma prior model [8.3.1.5.]
4. Reliability Data Analysis [8.4.]

1. How do you estimate life distribution parameters from censored
Data? [8.4.1.]
1. Graphical estimation [8.4.1.1.]
2. Maximum likelihood estimation [8.4.1.2.]
3. A Weibull maximum likelihood estimation example [8.4.1.3.]
2. How do you fit an acceleration model? [8.4.2.]
1. Graphical estimation [8.4.2.1.]
2. Maximum likelihood [8.4.2.2.]
3. Fitting models using degradation data instead of failures [8.4.2.3.]
3. How do you project reliability at use conditions? [8.4.3.]

4. How do you compare reliability between two or more populations? [8.4.4.]

5. How do you fit system repair rate models? [8.4.5.]
1. Constant repair rate (HPP/exponential) model [8.4.5.1.]
2. Power law (Duane) model [8.4.5.2.]
3. Exponential law model [8.4.5.3.]
6. How do you estimate reliability using the Bayesian gamma prior
model? [8.4.6.]
7. References For Chapter 8: Assessing Product Reliability [8.4.7.]

8.1. Introduction
8.1. Introduction
This section introduces the terminology and models that will be used to
describe and quantify product reliability. The terminology, probability
distributions and models used for reliability analysis differ in many
cases from those used in other statistical applications.
Detailed 1. Introduction
contents of 1. Why is the assessment and control of product reliability
Section 1
Important?
1. Quality versus reliability
2. Competitive driving factors
3. Safety and health considerations
2. What are the basic terms and models used for reliability
evaluation?
1. Repairable systems, non-repairable populations and
lifetime distribution models
2. Reliability or survival function
3. Failure (or hazard) rate
4. "Bathtub" curve
5. Repair rate or ROCOF
3. What are some common difficulties frequently found with
reliability data and how are they overcome?
1. Censoring
2. Lack of failures
4. What is "physical acceleration" and how do we model it?
5. What are some common acceleration models?
1. Arrhenius
2. Eyring
http://www.itl.nist.gov/div898/handbook/apr/section1/apr1.htm (1 of 2) [5/22/2001 1:27:37 PM]

8.1. Introduction
3. Other models
6. What are the basic lifetime distribution models used for
non-repairable populations?
1. Exponential
2. Weibull
3. Extreme value distributions
4. Lognormal
5. Gamma
6. Fatigue life (Birnbaum-Saunders)
7. Proportional hazards model
7. What are some basic repair rate models used for repairable
systems?
1. Homogeneous Poisson Process (HPP)
2. Non-Homogeneous Poisson Process (NHPP) with
power law
3. Exponential law
8. How can you evaluate reliability from the "bottoms - up"
(component failure mode to system failure rates)?
1. Competing risk model
2. Series model
3. Parallel or redundant model
4. R out of N model
5. Standby model
6. Complex systems
9. How can you model reliability growth?
1. NHPP power law
2. Duane plots
3. NHPP exponential law
10. How can Bayesian methodology be used for reliability
evaluation?

8.1.1. Why is the assessment and control of product reliability important?

8.1. Introduction
8.1.1. Why is the assessment and control of

product reliability important?
We depend In today's technological world nearly everyone depends upon the
on, demand, continued functioning of a wide array of complex machinery and
and expect equipment for their everyday health, safety, mobility and economic
reliable welfare. We expect our cars, computers, electric appliances, lights,
products. televisions, etc. to function whenever we need them, day after day , year
after year. When they fail the results can be catastrophic; injury, loss of
life and/or costly lawsuits can occur. More often, repeated failure leads
to annoyance, inconvenience and a lasting customer dissatisfaction that
can play havoc with the responsible company's marketplace position.
Shipping It takes a long time for a company to build up a reputation for reliability,
unreliable and only a short time to be branded as "unreliable" after shipping a
products flawed product. Continual assessment of new product reliability, and
can destroy on-going control of the reliability of everything shipped, is a critical
a company's necessity in today's competitive business arena.
reputation.
http://www.itl.nist.gov/div898/handbook/apr/section1/apr11.htm [5/22/2001 1:27:38 PM]

8.1.1.1. Quality versus reliability

8.1. Introduction
8.1.1.1. Quality versus reliability

Reliability is The everyday usage term "quality of a product" is loosely taken to
"quality mean its inherent degree of excellence. In industry, this is made more
changing precise by defining quality to be "conformance to requirements at the
over time". start of use". Assuming the product specifications adequately capture
customer requirements, the quality level can now be precisely
measured by the fraction of units shipped that meet specifications.
A motion But how many of these units still meet specifications after a week of
picture operation? Or after a month, or at the end of a one year warrantee
instead of a period? That is where "reliability" comes in. Quality is a snapshot at the
snapshot. start of life and reliability is a motion picture of the day by day
operation. Time zero defects are manufacturing mistakes that escaped
final test. The additional defects that appear over time are "reliability
defects" or reliability fallout.
Life The quality level can be described by a single fraction defective. To

distributions describe reliability fallout a probability model that describes the
model fraction fallout over time is needed. The is known as the life
fraction distribution model.
fallout over
time.

8.1.1.2. Competitive driving factors

8.1. Introduction
8.1.1.2. Competitive driving factors

Reliability is Accurate prediction and control of reliability plays an important role in
a major the profitability of a product. Service costs, for products within the
economic warrantee period or under a service contract, are a major expense and a
factor in significant pricing factor. Proper spare part stocking and support
determining a personnel hiring and training also depend upon good reliability fallout
product's predictions. On the other hand, missing reliability targets may invoke
success. contractual penalties and cost future business.
Companies that can economically design and market products that
meet their customer's reliability expectations have a strong competitive
advantage in today's marketplace.

8.1.1.3. Safety and health considerations

8.1. Introduction
8.1.1.3. Safety and health considerations

Some failures Sometimes equipment failure can have a major impact on human
have serious safety and/or health. Automobiles, planes, life support equipment,
social and power generating plants are a few examples.
consequences
and this should From the point of view of "assessing product reliability" we treat
be taken into these kinds of catastrophic failures no differently from the failure
account when when a key parameter measured on a manufacturing tool drifts
planning slightly out of specification, calling for an unscheduled maintenance
reliability action.
studies It is up to the reliability engineer (and the relevant customer) to
define what constitutes a failure in any reliability study. More
resource (test time and test units) should be planned for when an
incorrect reliability assessment could negatively impact safety and/or
health.

8.1.2. What are the basic terms and models used for reliability evaluation?

8.1. Introduction
8.1.2. What are the basic terms and models

used for reliability evaluation?
Reliability Reliability theory developed apart from the mainstream of probability
methods and and statistics, primarily as a tool to help nineteenth century maritime
terminology and life insurance companies compute profitable rates to charge their
began with customers. Even today the term "failure rate" and "hazard rate" are often
19th century used interchangeably.
insurance
companies The following sections will define some of the concepts, terms, and
models we need to describe, estimate and predict reliability

8.1.2.1. Repairable systems and non-repairable populations - lifetime distributionmodels

8.1. Introduction
8.1.2.1. Repairable systems and non-repairable

populations - lifetime distributionmodels
Life A repairable system is one which can be restored to satisfactory operation by any action,
distribution including parts replacements or changes to adjustable settings.
models
describe A non repairable population is one where individual items that fail are removed
how non permanently from the population. While failures may be replaced by units from either a
repairable similar or a different population, the members of the original population dwindle over
populations time until all have eventually failed.
fail over We begin with models and definitions for non repairable populations. Repair rates for
time repairable populations will be defined in a later section.
The theoretical population models used to describe unit lifetimes are known as Lifetime
Distribution Models. The population is generally considered to be all the possible unit
lifetimes of all the units that could be manufactured based on a particular design and
choice of materials and manufacturing process. A random sample of size n from this
population is the collection of failure times observed for a randomly selected group of n
units.
Any A lifetime distribution model can be any probability density function (or PDF) f(t)
continuous defined over the range of time from t = 0 to t = infinity. The corresponding cumulative
PDF distribution function (or CDF) F(t) is a very useful function, as it gives the probability
defined for that a randomly selected unit will fail by time t. The figure below shows the relationship
non native between f(t) and F(t) and gives three descriptions of F(t).
time can be
a lifetime
distribution
model

1. F(t) = the area under the PDF f(t) to the left of t.

2. F(t) = the probability that a single randomly chosen new
unit will fail by time t.
3. F(t) = the proportion of the entire population that fails
by time t.
The figure above also shows a shaded area under f(t) between the two times t1 and t2.
This area is [F(t2) - F(t1)] and represents the proportion of the population that fails
between times t1 and t2 (or the probability that a brand new randomly chosen unit will
survive to time t1 but fail before time t2).
Note that the PDF f(t) has only non-negative values and eventually either becomes 0 or
decreases towards 0. The CDF F(t) is monotonically increasing and goes from 0 to 1 as t
approaches infinity. In other words, the total area under the curve is always 1.
The The Weibull 2-parameter distribution is an example of a popular F(t). It has the CDF and
Weibull PDF equations given by:
model is a
good
example of
a life
distribution
where γ is the "shape" parameter and α is a scale parameter called the characteristic
life.
Example: A company produces automotive fuel pumps that fail according to a Weibull
life distribution model with shape parameter γ = 1.5 and scale parameter 8,000 (time
measured in use hours). If a typical pump is used 800 hours a year, what proportion are
likely to fail within 5 years?

Dataplot Solution: The Dataplot commands for the Weibull are:

Weibull
CDF SET MINMAX = 1
commands LET Y = WEICDF(((800*5)/8000),1.5)
and Dataplot computes Y to be .298 or about 30% of the pumps will fail in the first 5
years.

8.1.2.2. Reliability or survival function

8.1. Introduction
8.1.2.2. Reliability or survival function

Survival is the The Reliability Function R(t), also known as the Survival Function
complementary S(t), is defined by:
event to failure
R(t) = S(t) = the probability a unit survives beyond time t.
Since a unit either fails, or survives, and one of these two mutually
exclusive alternatives must occur, we have
R(t) = 1 - F(t), F(t) = 1 - R(t)
Calculations using R(t) often occur when building up from single
components to subsystems with many components. For example, if
one microprocessor comes from a population with reliability
function Rm(t) and two of them are used for the CPU in a system,
then the system CPU has a reliability function given by
Rcpu(t) = Rm2(t)
The reliability since both must survive in order for the system to survive. This
of the system is building up to the system from the individual components will be
the product of discussed in detail when we look at the "Bottoms - Up" method. The
the reliability general rule is: to calculate the reliability of a system of independent
functions of the components, multiply the reliability functions of all the components
components together.

8.1.2.3. Failure (or hazard) rate

8.1. Introduction

The The failure rate is defined as the (instantaneous) rate of failure for the
failure survivors to time t during the next instant of time. It is a rate per unit of
rate is the time similar in meaning to reading a car speedometer at a particular instant
rate at and seeing 45 mph. The next instant the failure rate may change and the
which the units that have already failed play no further role since only the survivors
population count.
survivors
at any The failure rate (or hazard rate) is denoted by h(t) and calculated from
given
instant are
"falling
over the
The failure rate is sometimes called a "conditional failure rate", since the
cliff"
denominator 1 - F(t) (i.e. the population survivors) converts the expression
into a conditional rate, given survival past time t.
Since h(t) is also equal to the negative derivative of lnR(t), we have the
useful identity:
If we let be the Cumulative Hazard Function then we

have F(t) = 1 - e-H(t). Two other useful identities that follow from these
formulas are:

It is also sometimes useful to define a single number average failure rate

over any interval (T1, T2) that"averages" the failure rate over that interval.
This rate, denoted by AFR(T1,T2), is a single number that can be used as a
specification or target for the population failure rate over that interval. If T1
is 0, it is dropped from the expression. Thus, for example, AFR(40,000)
would be the average failure rate for the population over the first 40,000
hours of operation.
The formulas for calculating AFR's are:

8.1.2.4. "Bathtub" curve

8.1. Introduction

A plot of
the
failure
rate
over
time for
most
If enough units from a given population are observed operating and failing over time, it is
products
relatively easy to compute week by week (or month by month) estimates of the failure rate
yields a
h(t). For example, if N12 units survive to start the 13th month of life, and r13 of them fail
curve
that during the next month (or 720 hours) of life, then a simple empirical estimate of h(t) averaged
looks across the 13th month of life (or between 8640 hours and 9360 hours of age), is given by (r13
like a / N12 * 720). Similar estimates are discussed in detail in the section on Empirical Model
drawing Fitting.
of a
bathtub. Over many years, and across a wide variety of mechanical and electronic components and
systems, people have calculated empirical population failure rates as units age over time and
repeatedly obtained a graph such as shown below. Because of the shape of this failure rate
curve, it has become widely known as the "Bathtub" curve.
The initial region that begins at time zero when a customer first begins to use the product is
characterized by a high but rapidly decreasing failure rate. This region is known as the Early
Failure Period (also referred to as Infant Mortality Period, from the actuarial origins of the
first bathtub curve plots). This decreasing failure rate typically lasts several weeks to a few
months.
Next, the failure rate levels off and remains roughly constant for (hopefully) the majority of
the useful life of the product. This long period of a level failure rate is known as the Intrinsic
Failure Period (also called the Stable Failure Period) and the constant failure rate level is
called the Intrinsic Failure Rate. Note that most systems spend most of their lifetimes
operating in this flat portion of the bathtub curve
Finally, if units from the population remain in use long enough, the failure rate begins to
increase as materials whereat and degradation failures occur at an ever increasing rate. This is
the Wear Out Failure Period .

The Bathtub Curve also applies (based on much empirical evidence) to Repairable Systems.
In this case, the vertical axis is the Repair Rate or the Rate of Occurrence of Failures
(ROCOF).

8.1.2.5. Repair rate or ROCOF

8.1. Introduction
8.1.2.5. Repair rate or ROCOF

Repair Rate A different approach is used for modeling repairable system repair rates.
models are The failures occur over time and the system is repaired to a state that
based on may be the same as new, or better, or worse. The frequency of repairs
counting the may be increasing, decreasing, or staying at a roughly constant rate.
cumulative
number of Let N(t) be a counting function that keeps track of the cumulative
failures over number of failures a given system has had from time zero to time t. N(t)
time. is a step function that jumps up one every time a failure occurs and stays
at the new level until the next failure.
Every system will have its own observed N(t) function over time. If we
observed the N(t) curves for a large number of similar systems, and
"averaged" these curves, we would have an estimate of M(t) = the
expected number (average number) of cumulative failures by time t for
these systems.
The Repair The derivative of M(t), denoted m(t), is known as the Repair Rate or
Rate (or the Rate Of Occurrence Of Failures at Time t or ROCOF.
ROCOF) is
the mean Models for N(t), M(t) and m(t) will be described in the section on Repair
rate of Rate Models.
failures per
unit time.

8.1.3. What are some common difficulties frequently found with reliability data and how are they overcome?

8.1. Introduction
8.1.3. What are some common difficulties

frequently found with reliability data
and how are they overcome?
The There are two closely related problems that are typical with reliability
Paradox of data, and not common with most other forms of statistical data. These
Reliability are:
Analysis: ● Censoring (when the observation period ends, not all units have
The more failed - some are survivors)
reliable a
● Lack of Failures (if there is too much censoring, even though a
product is,
the harder it large number of units may be under observation the information
is to get the in the data is limited due to the lack of actual failures)
failure data These problems cause considerable practical difficulty when planning
needed to reliability assessment tests and analyzing failure data. Some solutions
"prove" it is are discussed in the next two sections. Typically, the solutions involve
reliable! making additional assumptions and using complicated models.

8.1.3.1. Censoring

8.1. Introduction
8.1.3. What are some common difficulties frequently found with reliability data and how are they overcome?
8.1.3.1. Censoring
When not Consider a situation where we are reliability testing n (non repairable) units taken randomly
all units from a population. We are investigating the population to determine if its failure rate is
on test fail acceptable. In the typical test scenario, we have a fixed time T to run the units to see if they
we have survive or fail. The data obtained are called Censored Type I data.
censored
data Censored Type I Data
During the T hours of test we observe r failures (where r can be any number from 0 to n). The
(exact) failure times are t1, t2, ..., tr and there are (n - r) units that survived the entire T hour
test without failing. Note that T is fixed in advance and r is random, since we don't know how
many failures will occur until the test is run. Note also that we assume the exact times of
failure are recorded when there are failures.
This type of censoring is also called "right censored" data, since the times of failure to the
right (i.e. larger than T) are missing.
Another (much less common) way to test is to decide in advance you want to see exactly r
failure times and then test until they occur. For example, you might put 100 units on test and
decide you want to see at least half of them fail. Then r = 50, but T is unknown until the 50th
fail occurs. This is called Censored Type II data.
Censored Type II Data
We observe t1, t2, ..., tr, where r is specified in advance. The test ends at time T = tr, and (n-r)
units have survived. Again we assume it is possible to observe the exact time of failure for
failed units.
Type II censoring has the significant advantage that you know in advance how many failure
times your test will yield - this helps enormously when planning adequate tests. However, an
open-ended random test time is generally impractical from a management point of view and
this type of testing is rarely seen.

8.1.3.1. Censoring
Sometimes Readout or Interval Data

we don't
even know Sometimes exact times of failure are not known; only an interval of time in which the failure
the exact occurred is recorded. This kind of data is called Readout or Interval data and the situation is
time of shown in the figure below:
failure
.
Multicensored Data
In the most general case, every unit observed yields exactly one of the following three types
of information:
● a run-time if the unit did not fail while under observation
● an exact failure time
● an interval of time during which the unit failed.
The units may all have different run-times and/or readout intervals.
Many How do we handle censored data?

special
methods Many statistical methods can fit models and estimate failure rates even with censored data. In
have been later sections we will discuss the Kaplan-Meier approach, Probability Plotting, Hazard
developed Plotting, Graphical Estimation, and Maximum Likelihood Estimation.
to handle
Separating out Failure Modes
censored
data Note that when a data set consists of failure times that can be sorted into several different
failure modes, it is possible (and often necessary) to analyze and model each mode
separately. Consider all failures due to modes other than the one being analyzed as censoring
times, with the censored run-time equal to the time it failed due to the different (independent)
failure mode. This is discussed further in the competing risk section and later analysis
sections.

8.1.3.2. Lack of failures

8.1. Introduction
8.1.3. What are some common difficulties frequently found with reliability data and how are they
overcome?

Failure data When fitting models and estimating failure rates from reliability data,
is needed to the precision of the estimates (as measured by the width of confidence
accurately intervals) tends to vary inversely with the square root of the number of
assess and failures observed - not the number of units on test or the length of the
improve test. In other words, a test where 5 fail out of a total of 10 on test gives
reliability - more information than a test with 1000 units but only 2 failures.
this poses
problems Since the number of failures r is critical, and not the sample size n on
when testing test, it becomes increasingly difficult to assess the failure rates of highly
highly reliable components. Parts like memory chips, that in typical use have
reliable failure rates measured in parts per million per thousand hours, will have
parts few or no failures when tested for reasonable time periods with
affordable sample sizes. This gives little or no information for
accomplishing the two primary purposes of reliability testing, namely:
● accurately assessing population failure rates
● obtaining failure mode information to feedback for product

improvement.
Testing at How can tests be designed to overcome an expected lack of failures?

much higher
than typical The answer is to make failures occur by testing at much higher stresses
stresses can than the units would normally see in their intended application. This
yield creates a new problem: how can these over stresses failures be related to
failures but what would normally be expected to happen over the course of many
models are years at normal use stresses? The models that relate high stress
then needed reliability to normal use reliability are called acceleration models.
to relate
these back
to use stress


8.1.4. What is "physical acceleration" and how do we model it?

8.1. Introduction
8.1.4. What is "physical acceleration" and

how do we model it?
When Physical Acceleration (sometimes called True Acceleration or just
changing Acceleration) means that operating a unit at high stress (i.e. higher
stress is temperature or voltage or humidity or duty cycle, etc.) produces the
equivalent to same failures that would occur at typical use stresses, except that they
multiplying happen much quicker.
time to fail
by a Failure may be due to mechanical fatigue, corrosion, chemical reaction,
constant we diffusion, migration, etc. These are exactly the same events leading up
have true to failure occur at high stress as do at normal stress. Only the time scale
(physical) is changed.
acceleration
An When there is true acceleration, changing stress is equivalent to

Acceleration transforming the time scale used to record when failures occur. The
Factor is the transformations commonly used are linear, which means that time to
constant fail at high stress just has to be multiplied by a constant (the
multiplier acceleration factor) to get the equivalent time to fail at use stress.
between the
two stress Use the following notation:
levels ts = time to fail at stress tu = corresponding time to fail at use
Fs(t) = CDF at stress Fu(t) = CDF at use
fs(t) = PDF at stress fu(t) = PDF at use
hs(t) = failure rate at stress hu(t) = failure rate at use
Then, an acceleration factor AF between stress and use means the

following relationships hold:
Linear Acceleration Relationships
Time to Fail tu = AF × ts
Failure Probability Fu(t) = Fs(t/AF)
Reliability Ru(t) = Rs(t/AF)

8.1.4. What is "physical acceleration" and how do we model it?
PDF or Density Function fu(t) = (1/AF)fs(t/AF)

Failure Rate hu(t) = (1/AF) hs(t/AF)
Each failure Note: Acceleration requires there is a stress dependent physical process
mode has its causing change or degradation that leads to failure. Therefore, it is
own unlikely that a single acceleration factor will apply to more than one
acceleration failure mechanism. In general, different failure modes will be affected
factor. differently by stress and have different acceleration factors. Separate
out different types of failure when analyzing failure data.
Failure data
should be Also, a consequence of the linear acceleration relationships shown
separated by above (that follow directly from "true acceleration" is the following:
failure mode The Shape Parameter for the key life distribution models
when (Weibull, Lognormal) does not change for units operating
analyzed, if under different stresses. Plots on probability paper of data
acceleration from different stress cells will line up roughly parallel
is relevant.
These distributions, and probability plotting, will be discussed in later
Data from sections.
different
stress cells
have the
same slope
on
probability
paper (if
there is
acceleration)

8.1.5. What are some common acceleration models?

8.1. Introduction
8.1.5. What are some common acceleration

models?
Acceleration Acceleration factors show how time to fail at a particular operating
models stress level (for one failure mode or mechanism) can be used to predict
predict time the equivalent time to fail at a different operating stress level.
to fail as a
function of A model that predicts time to fail as a function of stress would be even
stress better than a collection of acceleration factors. If we write tf = G(S),
where G(S) is the model equation for an arbitrary stress level S, then
the acceleration factor between two stress levels S1 and S2 can be
evaluated simply by AF = G(S1)/G(S2). Now we can test at the higher
stress S2, obtain a sufficient number of failures to fit life distribution
models and evaluate failure rates, and use the Linear Acceleration
Relationships Table to predict what will occur at the lower use stress
S1.
A model that predicts time to fail as a function of operating stresses is

known as an acceleration model.
Acceleration Acceleration models are usually based on the physics or chemistry

models are underlying a particular failure mechanism. Successful empirical
often derived models often turn out to be approximations of complicated physics or
from physics kinetics models, when the theory of the failure mechanism is better
or kinetics understood. The following sections will consider a variety of powerful
models and useful models:
related to the ● Arrhenius
failure
mechanism. ● Eyring
● Other Models

8.1.5.1. Arrhenius

8.1. Introduction
8.1.5.1. Arrhenius
The One of the earliest and most successful acceleration models predicts
Arrhenius how time to fail varies with temperature. This empirically based model
model is known as the Arrhenius equation. It takes the form
predicts
failure
acceleration
due to
temperature
increase where T is temperature measured in degrees Kelvin (273.16 + degrees
Centigrade) at the point where the failure process takes place and k is
Boltzmann's constant (8.617 x 10-5 in eV/K). The constant A is a scaling
factor that drops out when calculating acceleration factors, while H
(pronounced "Delta H") is also called the activation energy, and is the
critical parameter in the model.
The The value of H depends on the failure mechanism and the materials
Arrhenius involved, and typically ranges between .3 or .4 up to 1.5 or even higher.
activation Acceleration factors between two temperatures increase exponentially
energy H as H increases.
is all you
need to The acceleration factor between a higher temperature T2 and a lower
know to temperature T1 is given by
calculate
temperature
acceleration
Using the value of k given above, this can be written in terms of T in

degrees Centigrade as

8.1.5.1. Arrhenius
Note that the only unknown parameter in this formula is H.

Example: The acceleration factor between 25°C and 125°C is 133 if
H = .5 and 17,597 if H = 1.0.
The Arrhenius model has been used successfully for failure mechanisms
that depend on chemical reactions, diffusion processes or migration
processes. This covers many of the non mechanical (or non material
fatigue) failure modes that cause electronic equipment failure.

8.1.5.2. Eyring

8.1. Introduction
8.1.5.2. Eyring
The Eyring
model has a Henry Eyring's contributions to chemical reaction rate theory have led
theoretical to a very general and powerful model for acceleration known as the
basis in Eyring Model. This model has several key features:
chemistry
● It has a theoretical basis from chemistry and quantum mechanics.
and quantum
mechanics ● If a chemical process (chemical reaction, diffusion, corrosion,
and can be migration, etc.) is causing degradation leading to failure, the

used to Eyring model describes how the rate of degradation varies with
model stress or, equivalently, how time to failure varies with stress.
acceleration ● The model includes temperature and can be expanded to include
when many other relevant stresses.
stresses are ● The temperature term by itself is very similar to the Arrhenius
involved empirical model, explaining why that model has been so
successful and establishing the connection between the H
parameter and the quantum theory concept of "activation energy
needed to cross an energy barrier and initiate a reaction".
The model for temperature and one additional stress takes the general
form:
where S1 could be some function of voltage or current or any other

relevant stress and the parameters , H, B, and C determine
acceleration between stress combinations. As with the Arrhenius Model,
k is Boltzmann's constant and temperature is in degrees Kelvin.
If we want to add an additional non thermal stress term the model
becomes

8.1.5.2. Eyring
and as many stresses as are relevant can be included by adding similar

terms.
Models with Note that the general Eyring model includes terms that have stress and
multiple temperature interactions (in other words, the effect of changing
stresses temperature varies, depending on the levels of other stresses). Most
generally models in actual use do not include any interaction terms, so that the
have no relative change in acceleration factors when only one stress changes
interaction does not depend on the level of the other stresses.
terms -
which means In models with no interaction, you can compute acceleration factors for
you can each stress, and multiply them together, This would not be true if the
multiply physical mechanism required interaction terms - but, at least to first
acceleration approximations, it seems to work for most examples in the literature.
factors due
to different
stresses
The Eyring Advantages of the Eyring Model

model can ● Can handle many stresses.
also be used
● Can be used to model degradation data as well as failure data.
to model
rate of ● The H parameter has a physical meaning and has been studied
degradation and estimated for many well known failure mechanisms and
leading to materials.
failure as a
function of
stress

8.1.5.2. Eyring
In practice, Disadvantages of the Eyring Model

the Eyring is ● Even with just two stresses there are 5 parameters to estimate.
usually too Each additional stress adds 2 more unknown parameters.
complicated
● Many of the parameters may have only a second order effect. For
to use in its
most general example, setting = 0 works quite well, since the temperature
form and term then becomes the same as in the Arrhenius model. Also, the
must be constants C and E are only needed if there is a significant
"customized" temperature interaction effect with respect to the other stresses.
or simplified ● The form in which the other stresses appear is not specified by
for any the general model and may vary according to failure mechanism.
particular In other words, S1 may be voltage or ln voltage or some other
failure function of voltage.
mechanism Many well known models are simplified versions of the Eyring model
with appropriate functions of relevant stresses chosen for S1 and S2.
Some of these will be shown in the Other Models section. The trick is to
find the right simplification to use for a particular failure mechanism.

8.1.5.3. Other models

8.1. Introduction

Many useful 1, This section will describe several acceleration models that have been
2 and 3 stress used successfully in the literature.
models are ● The (Inverse) Power Rule for Voltage
simple Eyring
models. Six ● The Exponential Voltage Model
are described ● Two Temperature/Voltage Models
● The Electromigration Model
● Three Stress Models (Temperature, Voltage and Humidity)
● The Coffin-Manson Mechanical Crack Growth Model
The (Inverse) Power Rule for Voltage
This model, used for capacitors, has only voltage dependency and
takes the form:
This a a very simplified Eyring model with , H, and C all 0, and S
= lnV, = -B.
The Exponential Voltage Model
In some cases, voltage dependence fits an exponential model better.
Two Temperature/Voltage Models

Temperature / Voltage models are common in the literature and take
one of the two forms given below:

Again, these are just simplified two stress Eyring models with the
appropriate choice of constants and functions of voltage.
The Electromigration Model
Electromigration is a semiconductor failure mechanism where open
failures occur in metal thin film conductors due to the movement of
ions toward the anode. This ionic movement is accelerated high
temperatures and high current density. The (modified Eyring) model
takes the form
wheren J is current density. H is typically between .5 and 1.2

electron volts, while an n around 2 is common.
Three-Stress Models (Temperature, Voltage and Humidity)
Humidity plays an important role in many failure mechanisms that
depend on corrosion or ionic movement. A common 3-stress model
takes the form
Here RH is percent relative humidity. Other obvious variations on

this model would be to use an exponential voltage term and/or an
exponential RH term.
Even this simplified Eyring 3-stress model has 4 unknown parameters
and it would require an extensive experimental setup to fit the model
and calculate acceleration factors.

The The Coffin-Manson Mechanical Crack Growth Model

Coffin-Manson
Model is a Models for mechanical failure, material fatigue or material
useful deformation are not forms of the Eyring model. These models
non-Eyring typically have terms relating to cycles of stress or frequency of use or
model for levels of change in temperatures. A model of this type known as the
crack growth (modified) Coffin-Manson model has been used successfully to model
or material crack growth in solder and other metals due to repeated temperature
fatigue cycling as equipment is turned on and off. This model takes the form
with
● Nf = the number of cycles to fail
● f = the cycling frequency
● T = the temperature range during a cycle
and G(Tmax) is an Arrhenius term evaluated at the maximum
temperature reached in each cycle.
Typical values for the cycling frequency exponent and the

temperature range exponent are around -1/3 and 2, respectively
(note that reducing the cycling frequency reduces the number of
cycles to failure). The H activation energy term in G(Tmax) is
around 1.25.

8.1.6. What are the basic lifetime distribution models used for non-repairable populations?

8.1. Introduction
8.1.6. What are the basic lifetime

distribution models used for
non-repairable populations?
A handful of There are a handful of parametric models that have successfully served
lifetime as population models for failure times arising from a wide range of
distribution products and failure mechanisms. Sometimes there are probabilistic
models have arguments based on the physics of the failure mode that tend to justify
enjoyed the choice of model. Other times the model is used solely because of its
great empirical success fitting actual failure data.
practical
success Seven models will be described in this section:
1. The Exponential
2. The Weibull
3. The Extreme Value
4. The Lognormal
5. The Gamma
6. The Birnbaum-Saunders Model
7. The Proportional hazards Model

8.1.6.1. Exponential

8.1. Introduction
● Formulas and Plots
● Uses of the Exponential Distribution Model
● DATAPLOT and EXCEL Functions for the Exponential
All the key Formulas and Plots

formulas
for using The exponential model, with only one unknown parameter, is the simplest of all life
the distribution models. The key equations for the exponential are shown below:
Exponential
model
(everything
you ever
wanted to
know but
were afraid
to find out)
Note that the failure rate reduces to the constant for any time. The exponential distribution
is the only distribution to have a constant failure rate. Also, another name for the exponential
mean is the Mean Time To Fail or MTTF and we have MTTF = 1/ .
The Cum Hazard function for the exponential is just the integral of the failure rate or H(t) =
t.
The PDF for the exponential has the familiar shape shown below.

The
Exponential
distribution
'shape'
The
Exponential
CDF
Below is an example of typical exponential lifetime data displayed in Histogram form with
corresponding exponential PDF drawn through the histogram.

Histogram
of
Exponential
Data
The Uses of the Exponential Distribution Model

Exponential 1. Because of its constant failure rate property, the exponential is an excellent model for
models the the long flat "intrinsic failure" portion of the Bathtub Curve Since most components
flat portion and systems spend most of their lifetimes in this portion of the Bathtub Curve, this
of the justifies frequent use of the exponential (when early failures or wear out is not a
"bathtub" concern).
curve -
where most 2. Just as it is often useful to approximate a curve by piecewise straight line segments, we
systems can approximate any failure rate curve by week by week or month by month constant
spend most rates that are the average of the actual changing rate during the respective time
of their duration's. That way we can approximate any model by piecewise exponential
'lives' distribution segments patched together.
3. Some natural phenomenon have a constant failure rate (or occurrence rate) property;
for example, the arrival rate of cosmic ray alpha particles or geiger counter tics. The
exponential model works well for inter arrival times (while the Poisson distribution
describes the total number of events in a given period). When these events trigger
failures, the exponential life distribution model will naturally apply.

Dataplot DATAPLOT and EXCEL Functions for the Exponential

and EXCEL
functions The Dataplot commands EXPPDF and EXPCDF calculate the exponential PDF and CDF for
for the the standardized case where = 1. To evaluate the PDF and CDF at 100 hours for an
Exponential
exponential with = .01, the commands would be
model
LET A = .01*EXPPDF(.01*100)
LET B = EXPCDF(.01*100)
and the response would be .003679 for the pdf and .63212 for the cdf.
Dataplot can do a probability plot of exponential data, normalized so that a perfect
exponential fit is a diagonal line with slope 1. The following commands generate 100 random
exponential observations ( = .01) and generate the probability plot that follows.
LET Y = EXPONENTIAL RANDOM NUMBERS FOR I = 1 1 100
LET Y = 100*Y
TITLE AUTOMATIC
X1LABEL THEORETICAL (NORMALIZED) VALUE;Y1LABEL DATA
VALUE;
EXPONENTIAL PROBABILITY PLOT Y
Dataplot
Exponential
probability
plot
EXCEL also has built in functions for the exponential PDF and CDF. The PDF is given by
EXPONDIST(x, ,false) and the CDF is given by EXPONDIST(x, ,true). Putting in 100 for
x and .01 for will give the same answers as given by Dataplot.

8.1.6.2. Weibull

8.1. Introduction
8.1.6.2. Weibull
● Uses of the Weibull Distribution Model
● DATAPLOT and EXCEL Functions for the Weibull
Weibull Formulas and Plots

Formulas
The Weibull is a very flexible life distribution model with two parameters. It has CDF
and PDF and other key formulas given by:
where is a scale parameter (the Characteristic Life) and (gamma) is known as the
Shape Parameter and is the Gamma function with (N) = (N-1)! for integer N.
The Cum Hazard function for the Weibull is the integral of the failure rate or

8.1.6.2. Weibull
A more general 3 parameter form of the Weibull includes an additional waiting time
parameter µ (sometimes called a shift or location parameter). The formulas for the 3
parameter Weibull are easily obtained from the above formulas by replacing t by (t - µ)
wherever t appears. No failure can occur before µ hours, so the time scale starts at µ, and
not 0. If a shift parameter µ is known (based, perhaps, on the physics of the failure mode)
then all you have to do is subtract µ from all the observed failure times and/or readout
times and analyze the resulting shifted data with a 2 parameter Weibull.
NOTE: Various texts and articles in the literature use a variety of different symbols for
the same Weibull parameters. For example, the characteristic life is sometimes called c
(or = nu or = eta) and the shape parameter is also called m (or = beta). To add to
the confusion, EXCEL calls the characteristic life and the shape and some authors
even parameterize the density function differently, using a scale parameter
Special Case: When = 1, the Weibull reduces to the Exponential Model, with = 1/
= the mean time to fail (MTTF).
Depending on the value of the shape parameter , the Weibull model can empirically fit
a wide range of data histogram shapes. This is shown by the PDF example curves below.
Weibull
data
'shapes'
From a failure rate model view point, the Weibull is a natural extension of the constant
failure rate exponential model, since the Weibull has a polynomial failure with exponent
{ - 1}. This makes all the failure rate curves shown in the following plot possible.

8.1.6.2. Weibull
Weibull
failure rate
'shapes'
The Weibull Uses of the Weibull Distribution Model

is very 1. Because of its flexible shape and ability to model a wide range of failure rates, the
flexible and Weibull has been used successfully in many applications as a purely empirical
also has model.
theoretical
2. The Weibull model can be derived theoretically as a form of Extreme Value
justification
in many Distribution, governing the time to occurrence of the "weakest link" of many
applications competing failure processes. This may explain why it has been so successful in
applications such as capacitor, ball bearing, relay and material strength failures
3. Another special case of the Weibull occurs when the shape parameter is 2. The
distribution is called the Rayleigh Distribution and it turns out to be the theoretical
probability model for the magnitude of radial error when the x and y coordinate
errors are independent normals with 0 mean and the same standard deviation.

8.1.6.2. Weibull
Dataplot DATAPLOT and EXCEL Functions for the Weibull

and EXCEL
functions The following commands in Dataplot will evaluate the PDF and CDF of a Weibull at
for the time T, with shape and characteristic life .
Weibull SET MINMAX 1
LET PDF = (1/ )*WEIPDF((T/ ),
LET CDF = WEICDF((T/ ), )
For example, if T = 1000 and = 1.5 and = 5000, the above commands will produce
a PDF of .000123 and a CDF of .08556.
NOTE: Whenever using Dataplot for Weibull analysis, you must start by setting
MINMAX equal to 1.
To generate Weibull random numbers from a Weibull with shape 1.5 and characteristic
life 5000 use the following commands:
SET MINMAX 1
LET GAMMA = 1.5
LET SAMPLE = WEIBULL RANDOM NUMBERS FOR I = 1 1 100
LET SAMPLE = 5000*SAMPLE
Next, to see how well these "random Weibull data points" actually fit a Weibull, we plot
them on "Weibull" paper to check whether they line up following a straight line. The
commands (following the last commands above) are:
X1LABEL LOG TIME;Y1LABEL CUM PROBABILITY
PLOT WEIBULL SAMPLE
The resulting plot is below. Note the log scale used is base 10
Dataplot
Weibull
Probability
Plot

8.1.6.2. Weibull
EXCEL also has Weibull CDF and PDF built in functions. EXCEL calls the shape
parameter = alpha and the characteristic life = beta. The following command
evaluates the Weibull PDF for time 1000 when the shape is 1.5 and the characteristic life
is 5000:
WEIBULL(1000,1.5,5000,FALSE)
For the corresponding CDF
WEIBULL(1000,1.5,5000,TRUE)
The returned values (.000123 and .085559, respectively) are the same as calculated by
Dataplot.

8.1.6.3. Extreme value distributions

8.1. Introduction

● Description, Formulas and Plots
● Uses of the Extreme Value Distribution Model
● DATAPLOT Functions for the Extreme Value Distribution
The Extreme Description, Formulas and Plots

Value
Distribution We have already referred to Extreme Value Distributions when describing the uses of the
usually Weibull distribution. Extreme value distributions are the limiting distributions for the
refers to the minimum or the maximum of a very large collection of random observations from the same
distribution arbitrary distribution. Gumbel (1958) showed that for any well behaved initial distribution
of the (F(x) is continuous and has an inverse), only a few models are needed, depending on whether
minimum of you are interested in the maximum or the minimum and also if the observations are bounded
a large above or below.
number of
In the context of reliability modeling, extreme value distributions for the minimum are
unbounded
frequently encountered. For example, if a system consists of n identical components in series,
random
and the system fails when the first of these components fails, then system failure times are the
observations
minimum of n random component failure times. Extreme value theory says that, independent
of the choice of component model, the system model will approach a Weibull as n gets large.
The same reasoning can also be applied at a component level, if the component failure occurs
when the first of many similar competing failure processes reaches a critical level.
The distribution often referred to as the Extreme Value Distribution (Type I) is the limiting
distribution of the minimum of a large number of unbounded identically distributed random
variables. The PDF and CDF are given by:
Extreme
Value
Distribution
formulas
and PDF
shapes
If the x values are bounded below (as is the case with times of failure) then the limiting
distribution is the Weibull. Formulas and uses of the Weibull have already been discussed.
PDF Shapes for the (minimum) Extreme Value Distribution (Type I) are shown in the
following figure.

The natural Uses of the Extreme Value Distribution Model

log of 1. In any modeling application where the variable of interest is the minimum of many
extreme random factors, all of which can take positive or negative values, try the extreme value
value distribution as a likely candidate model. For lifetime distribution modeling, since failure
distribution times are bounded below by zero, the Weibull distribution is a better choice.
data is
2. The Weibull distribution and the extreme value distribution have a useful mathematical
Weibull
relationship. If t1, t2, ...,tn are a sample of random times of fail from a Weibull
data
distribution, then ln t1, ln t2, ...,ln tn are random observations from the extreme value
distribution. In other words, the natural log of a Weibull random time is an extreme
value random observation.
Because of this relationship, computer programs and graph papers designed for the
extreme value distribution, can be used to analyze Weibull data. The situation exactly
parallels using normal distribution programs to analyze lognormal data, after first taking
natural logarithms of the data points.

Dataplot DATAPLOT for the Extreme Value Distribution

commands
for the Assume µ = ln 200,000 = 12.206 and = 1/2 = .5. The extreme value distribution with these
extreme parameters could be obtained by taking natural logarithms of data from a Weibull population
value
with characteristic life = 200,000 and shape = 2. We will use Dataplot to evaluate PDF's,
distribution
CDF's and generate random numbers from this distribution. Note that you must first set
MINMAX to 1, in order to do (minimum) extreme value type I calculations.
SET MINMAX 1
LET BET = .5
LET M = LOG(200000)
LET X = DATA 5 8 10 12 12.8
LET PD = (1/BET)*EV1PDF((X-M)/BET)
LET CD = EV1CDF((X-M)/BET)
Dataplot will calculate PDF and CDF values corresponding to the points 5, 8, 10, 12, 12.8. The
PDF's are .110E-5, .444E-3, .024, .683 and .247. The CDF's are .551E-6, .222E-3, .012, .484
and .962.
Finally, we generate 100 random numbers from this distribution and construct an extreme
value distribution probability plot as follows:
LET SAM = EXTREME VALUE TYPE 1 RANDOM NUMBERS FOR I = 1 1
100
LET SAM = (BET*SAMPLE) + M
EXTREME VALUE TYPE 1 PROBABILITY PLOT SAM
Data from an extreme value distribution will line up approximately along a straight line when
this kind of plot is done. The slope of the line is an estimate of , and the "y-axis" value on
the line corresponding to the "x-axis" 0 point is an estimate of µ. For the graph above, these
turn out to be very close to the actual values of and µ.


8.1.6.4. Lognormal

8.1. Introduction
8.1.6.4. Lognormal
● Uses of the Lognormal Distribution Model
● DATAPLOT and EXCEL Functions for the Lognormal
Lognormal Formulas and Plots

Formulas and
relationship The lognormal life distribution, like the Weibull, is a very flexible model that can empirically
to the normal fit many types of failure data. The two parameter form has parameters = the shape
distribution parameter and T50= the median (a scale parameter).
Note: If time to failure, tf, has a lognormal distribution, then the (natural) logarithm of time to
failure has a normal distribution with mean µ = ln T50 and standard deviation . This makes
lognormal data convenient to work with; just take natural logarithms of all the failure times and
censoring times and analyze the resulting normal data. Later on, convert back to real time and
lognormal parameters using as the lognormal shape and T50 = eµ as the (median) scale
parameter.
Below is a summary of the key formula's for the lognormal.

8.1.6.4. Lognormal
Note: A more general 3 parameter form of the lognormal includes an additional waiting time
parameter (sometimes called a shift or location parameter). The formulas for the 3 parameter
lognormal are easily obtained from the above formulas by replacing t by (t - ) wherever t
appears. No failure can occur before hours, so the time scale starts at and not 0. If a shift
parameter is known (based, perhaps, on the physics of the failure mode), then all you have to
do is subtract from all the observed failure times and/or readout times and analyze the
resulting shifted data with a 2 parameter lognormal.
Examples of lognormal PDF and failure rate plots are shown below. Note that lognormal shapes
for small sigma's are very similar to Weibull shapes when the shape parameter is large and
large sigma's give plots similar to small Weibull 's. Both distributions are very flexible and it
is often difficult to choose which to use based on empirical fits to small samples of (possibly
censored) data.
Lognormal
data 'shapes'

8.1.6.4. Lognormal
Lognormal
failure rate
'shapes'
A very Uses of the Lognormal Distribution Model

flexible model 1. As shown in the preceding plots, the lognormal PDF and failure rate shapes are flexible
that also can enough to make it a very useful empirical model. In addition, the relationship to the
apply normal (just take natural logarithms of all the data and time points and you have "normal"
(theoretically) data) makes it easy to work with mathematically, with many good software analysis
to many programs available to treat normal data.
degradation
2. The lognormal model can be theoretically derived under assumptions matching many
process
failure degradation processes common to electronic (semi-conductor) failure
failure modes
mechanisms. Some of these are: corrosion, diffusion, migration, crack growth,
electromigration, and, in general, failures resulting from chemical reactions or processes.
That does not mean that the lognormal is always the correct model for these mechanisms,
but it does perhaps explain why it has been empirically successful in so many of these
cases.
A brief sketch of the theoretical arguments leading to a lognormal model follows.

Applying the Central Limit Theorem to small additive errors in the log
domain and justifying a normal model is equivalent to justifying the
lognormal model in real time when a process moves towards failure based
on the cumulative effect of many small "multiplicative" shocks. More
precisely, if at any instant in time a degradation process undergoes a small
increase in the total amount of degradation that is proportional to the current
total amount of degradation, then it is reasonable to expect the time to failure
(i.e. reaching a critical amount of degradation) to follow a lognormal
distribution (Kolmogorov, 1941).
A more detailed description of the multiplicative degradation argument appears in a later
section.

8.1.6.4. Lognormal
Dataplot and DATAPLOT and EXCEL Functions for the Lognormal

EXCEl
lognormal The following commands in Dataplot will evaluate the PDF and CDF of a lognormal at time T,
functions with shape and median life (scale parameter) T50:
LET PDF = (1/ T50)*LGNPDF((T/T50), )
LET CDF = LGNCDF((T/T50), )
For example, if T = 5000 and = .5 and T50 = 20,000, the above commands will produce a
PDF of .34175E-5 and a CDF of .002781 and a failure rate of PDF/(1-CDF) = .3427 %/K
To generate 100 lognormal random numbers from a lognormal with shape .5 and median life
20,000 use the following commands:
LET SAMPLE = LOGNORMAL RANDOM NUMBERS FOR I = 1 1 100
LET SAMPLE = 20,000*(SAMPLE**.5)
Next, to see how well these "random" lognormal data points" actually fit a lognormal, we plot
them using the lognormal plot command. First we have to set = SD to .5 (see PPCC PLOT
for how to estimate the value of SD from actual data).
LET SD = .5
X1LABEL EXPECTED (NORMALIZED) VALUES;Y1LABEL TIME
LOGNORMAL PROBABILITY PLOT SAMPLE
The resulting plot is below. Points that line up approximately on a straight line indicates a good
fit to a lognormal (with shape SD = .5). The time that corresponds to the (normalized) x-axis
T50of 1 is the estimated T50 according to the data In this case it is close to 20,000, as expected.
Dataplot
lognormal
probability
plot

8.1.6.4. Lognormal
Finally, we note that EXCEL has a built in function to calculate the lognormal CDF. The
command is =LOGNORMDIST(5000,9.903487553,0.5) to evaluate the CDF of a lognormal at
time T = 5000 with = .5 and T50 = 20,000 and ln T50 = 9.903487553. The answer returned is
.002781. There is no lognormal PDF function in EXCEL. The normal PDF can be used as
follows:
=(1/5000)*NORMDIST(8.517193191,9.903487553,0.5,FALSE)
where 8.517193191 is ln 5000 and "FALSE" is needed to get PDF's instead of CDF's. The
answer returned is 3.42E-06.

8.1.6.5. Gamma

8.1. Introduction
8.1.6.5. Gamma
● Uses of the Gamma Distribution Model
● DATAPLOT and EXCEL Functions for the Gamma
Formulas Formulas and Plots

for the
gamma There are two ways of writing (parameterizing) the gamma distribution that are common in the
model literature. In addition, different authors use different symbols for the shape and scale parameters.
Below we show three ways of writing the gamma, with a = = the "shape" parameter and b
=1/ the scale parameter. The first choice of parameters (a,b) will be the most convenient for
later applications of the gamma. EXCEL uses while Dataplot uses

8.1.6.5. Gamma
The
Note: When a = 1, the gamma reduces to a exponential distribution with b = .
exponential
is a special Another well known statistical distribution, the Chi Square, is also a special case of the gamma. A
case of the Chi Square distribution with n degrees of freedom is the same as a gamma with a = n/2 and b = .5
gamma (or = 2).
The following plots give examples of gamma PDF, CDF and failure rate shapes.
Shapes for
Gamma
data
Gamma
CDF
shapes

8.1.6.5. Gamma
Gamma
failure rate
shapes

8.1.6.5. Gamma
The Uses of the Gamma Distribution Model

gamma is 1. The gamma is a flexible life distribution model that may offer a good fit to some sets of
used in failure data. It is not, however, widely used as a life distribution model for common failure
"Standby" mechanisms.
system
2. The gamma does arise naturally as the time to first fail distribution for a system with
models and
standby exponentially distributed backups. If there are n-1 standby backup units, and the
also for
Bayesian system and all backups have exponential lifetimes with parameter , then the total lifetime
reliability
has a gamma distribution with a = n and b = . Note: when a is a positive integer, the
analysis
gamma is sometimes called an Erlang distribution. The Erlang distribution is used
frequently in queuing theory applications.
3. A common use of the gamma model occurs in Bayesian reliability applications. When a
system follows an HPP (exponential) model with a constant repair rate , and it is desired
to make use of prior information about possible values of , a gamma Bayesian prior for
is a convenient and popular choice.
Dataplot Dataplot and EXCEL Functions for the Gamma

and
EXCEL To calculate the PDF, CDF, Reliability and failure rate at time t for a gamma with parameters a
gamma and b = 1/ use the following Dataplot statements:
functions LET PDF = b*GAMPDF((t*b),a)
LET CDF = GAMCDF((t*b),a)
LET REL = 1-CDF
LET FR = PDF/REL
Using an example solved in the section on standby models, if a = 2 and b = 1/30 and t is 24
months the statements are:
LET PDF = (1/30)*GAMPDF((24/30),2) response is .01198
LET CDF = GAMCDF((24/30),2) response is .1912
LET REL = 1-CDF response is .8088
LET FR=PDF/REL response is .0148
To generate random gamma data we first have to set the "a" parameter (called "gamma" by
Dataplot). The following commands generate 100 gamma data points chosen randomly from a
gamma distribution with parameters a and b:
LET GAMMA = a
LET DATA = GAMMA RANDOM NUMBERS FOR I = 1 1 100
LET DATA = (1/b)*DATA
For the above example this becomes
LET GAMMA = 2
LET DATA = GAMMA RANDOM NUMBERS FOR I = 1 1 100
LET DATA = 30*DATA
Continuing this example, we can now do a gamma probability plot of the 100 points in DATA.
The commands are
LET GAMMA = 2
X1LABEL EXPECTED (NORMALIZED) VALUES;Y1LABEL TIME
GAMMA PROBABILITY PLOT DATA
The resulting plot is shown below.

8.1.6.5. Gamma
EXCEL also has built in functions to evaluate the gamma pdf and cdf. The syntax is:
=GAMMADIST(t,a,1/b,FALSE) for the PDF
=GAMMADIST(t,a,1/b,TRUE) for the CDF

8.1.6.6. Fatigue life (Birnbaum-Saunders)

8.1. Introduction

A model In 1969, Birnbaum and Saunders described a life distribution model that could be derived
based on from a physical fatigue process where crack growth causes failure. Since one of the best ways
cycles of to choose a life distribution model is to derive it from a physical/statistical argument that is
stress consistent with the failure mechanism, the Birnbaum-Saunders Fatigue Life Distribution is
causing worth considering.
degradation ● Formulas and Plots for the Birnbaum-Saunders Model
or crack
growth ● Derivation and Use of the Birnbaum-Saunders Model
● Dataplot Functions for the Birnbaum-Saunders Model
Formulas and Plots for the Birnbaum-Saunders Model
Formulas
and shapes
for the
Fatigue
Life model
The PDF, CDF, mean and variance for the Birnbaum-Saunders Distribution are shown below.
The parameters are: = a shape parameter; µ a scale parameter. These are the parameters
used in Dataplot, but there are other choices also common in the literature (see the parameters
used for the derivation of the model).

PDF shapes for the model vary from highly skewed and long tailed (small gamma values) to
nearly symmetric and short tailed as gamma increases. This is shown in the figure below.

Corresponding failure rate curves are shown in the next figure.
If crack Derivation and Use of the Birnbaum-Saunders Model:

growth in
each stress Consider a material that continually undergoes cycles of stress loads. During each cycle, a
cycle is a dominant crack grows towards a critical length that will cause failure. Under repeated
random application of n cycles of loads, the total extension of the dominant crack can be written as
amount
independent
of past
cycles of where we assume the Yj are independent and identically distributed non-negative random
growth, the
Fatigue variables with mean µ and variance . Suppose failure occurs at the N-th cycle, when wn
Life mode
first exceeds a constant critical value w. If n is large we can use a central limit theorem
model may
argument to conclude that
apply.
Since there are many cycles, each lasting a very short time, we can replace the discrete
number of cycles N needed to reach failure by the continuous time tf needed to reach failure.
The cdf F(t) of tf is given by

Here denotes the standard normal cdf. The parameters and are an alternative way of
writing the Birnbaum-Saunders distribution that is often used ( , as compared to
the way the formulas were parameterized earlier in this section).
Note:
The critical assumption in the derivation, from a physical point of view, is that the crack
growth during any one cycle is independent of the growth during any other cycle. Also, the
growth has approximately the same random distribution, from cycle to cycle. This is a very
different situation from the proportional degradation argument used to derive a log normal
distribution model, where the rate of degradation at any point in time depends on the total
amount of degradation that has occurred up to that time.
This kind of The Birnbaum-Saunders assumption, while physically restrictive, is consistent with a
physical deterministic model from materials physics known as Miner's Rule (Miner's Rule implies that
degradation the damage that occurs after n cycles, at a stress level that produces a fatigue life of N cycles,
is is proportional to n/N). So, when the physics of failure suggests Miner's Rule applies, the
consistent Birnbaum-Saunders model is a reasonable choice for a life distribution model.
with
Miner's
Rule.
Dataplot
commands Dataplot Functions for the Birnbaum-Saunders Model
for the
Fatigue The PDF for a Birnbaum-Saunders (Fatigue Life) distribution with parameters µ, is
Life model evaluated at time t by:
LET PDF = (1/µ*FLPDF((t/µ, ).
The CDF is
LET CDF = FLCDF((t/µ), ).
To generate 100 random numbers, when µ = 5000, = 2, for example, type the following
Dataplot commands:
LET GAMMA = 2
LET DATA = FATIGUE LIFE RANDOM NUMBERS FOR
I = 1 1 100
LET DATA = 5000*DATA
Finally, we can do a Fatigue Life Probability Plot of the 100 data points in DATA by
LET GAMMA = 2
FATIGUE LIFE PROBABILITY PLOT DATA
and the points on the resulting plot (shown below) line up roughly on a straight line, as
expected for data correctly modeled by the Birnbaum-Saunders distribution.

Notes
1. We set GAMMA equal to 2 before doing the probability plot because we knew its
value. If we had real experimental data (with no censoring), first we would run PPCC
to estimate gamma. The command is: FATIGUE LIFE PPCC PLOT DATA. To see the
estimated value of gamma we would type WRITE SHAPE. Then, we would type LET
GAMMA = SHAPE before running the Fatigue Life Probability Plot.
2. The slope of the line through the points on the probability plot is an estimate of the
scale parameter µ.

8.1.6.7. Proportional hazards model

8.1. Introduction

The The proportional hazards model, proposed by Cox (1972), has been
proportional used primarily in medical testing analysis, to model the effect of
hazards secondary variables on survival. It is more like an acceleration model
model is often than a specific life distribution model, and its strength lies in its ability
used in to model and test many inferences about survival without making any
survival specific assumptions about the form of the life distribution model.
analysis
(medical This section will give only a brief description of the proportional
testing) hazards model, since it has limited engineering applications.
studies. It is Proportional Hazards Model Assumption
not used
much with Let z = {x, y, ...} be a vector of 1 or more explanatory variables
engineering believed to affect lifetime. These variables may be continuous (like
data. temperature in engineering studies, or dosage level of a particular drug
in medical studies) or they may be indicator variables with the value 1,
if a given factor or condition is present, and 0 otherwise.
Let the hazard rate for a nominal (or baseline) set z0 = (x0,y0, ...) of
these variables be given by h0(t), where h0(t) is a legitimate hazard
function (failure rate) for some unspecified life distribution model.

The The proportional hazards model assumes we can write the changed
proportional hazard function for a new value of z as
hazard model
assumes hz(t) = g(z)h0(t)
changing a In other words, changing z, the explanatory variable vector, results in a
stress new hazard function that is proportional to the nominal hazard
variable (or function, and the proportionality constant is a function of z, g(z),
explanatory independent of the time variable t.
variable) has
the effect of A common and useful form for f(z) is the Log Linear Model which
multiplying has the equation: g(x) = eax for one variable, g(x,y) = eax+by for two
the hazard variables, etc.
rate by a
constant. Properties and Applications of the Proportional Hazards Model
1. The proportional hazards model is equivalent to the acceleration
factor concept if and only if the life distribution model is a
Weibull (which includes the exponential model, as a special
case). For a Weibull with shape parameter , and an
acceleration factor AF between nominal use fail time t0 and high
stress fail time ts (with t0 = AFts) we have g(s) = AF . In other
words, hs(t) = AF h0(t).
2. Under a log linear model assumption for g(z), without any
further assumptions about the life distribution model, it is
possible to analyze experimental data and compute maximum
likelihood estimates and use likelihood ratio tests to determine
which explanatory variables are highly significant. In order to do
this kind of analysis, however, special software is needed.
More details on the theory and applications of the proportional hazards
model may be found in Cox and Oakes (1984).

8.1.7. What are some basic repair rate models used for repairable systems?

8.1. Introduction
8.1.7. What are some basic repair rate

models used for repairable systems?
Models for N(t), M(t) and m(t) were defined in the section on Repair Rates. Repair
repair rates rate models are defined by first picking a functional form for M(t), the
of expected number of cumulative failures by time t. Taking the derivative
repairable of this gives the repair rate model m(t).
systems
In the next three sections we will describe three models, of increasing
complexity, for M(t). They are: the Homogeneous Poisson Process; the
Non Homogeneous Poisson Process following a Power law and the Non
Homogeneous Poisson Process following an Exponential law.

8.1.7.1. Homogeneous Poisson Process (HPP)

8.1. Introduction
8.1.7.1. Homogeneous Poisson Process

(HPP)
Repair rate
(ROCOF)
models and
formulas
The simplest useful model for M(t) is M(t) = t and the repair rate (or
ROCOF) is the constant m(t) = . This model comes about when the
interarrival times between failures are independent and identically
distributed according to the exponential distribution, with parameter .
This basic model is also known as a Homogeneous Poisson Process
(HPP). The following formulas apply:

HPP model Despite the simplicity of this model, it is widely used for repairable
fits flat equipment and systems throughout industry. Justification for this comes,
portion of in part, from the shape of the empirical Bathtub Curve. Most systems (or
"bathtub" complex tools or equipment) spend most of their "lifetimes" operating in
curve the long flat constant repair rate portion of the Bathtub Curve. The HPP
is the only model that applies to that portion of the curve, so it is the
most popular model for system reliability evaluation and reliability test
planning.
Planning reliability assessment tests (under the HPP assumption) is
covered in a later section, as is estimating the MTBF from system
failure data and calculating upper and lower confidence limits.

Poisson Note that in the HPP model, the probability of having exactly k failures
relationship
by time T is given by the Poisson distribution with mean T (see
and
formula for P(N(T) = k) above). This can be evaluated by the Dataplot
Dataplot
expression:
and EXCEL
functions
LET Y = POIPDF(k, T)
or by the EXCEL expression:
POISSON(k, T, FALSE)

8.1.7.2. Non-Homogeneous Poisson Process (NHPP) - power law

8.1. Introduction
8.1.7.2. Non-Homogeneous Poisson

Process (NHPP) - power law
The repair A flexible model (that has been very successful in many applications)
rate for a for the expected number of failures in the first t hours, M(t), is given by
NHPP the polynomial
following the
Power law
The repair rate (or ROCOF) for this model is
The Power
The HPP model has a the constant repair rate m(t) = . If we substitute
law model is
very flexible an arbitrary function (t) for , we have a Non Homogeneous
and contains
Poisson Process (NHPP) with Intensity Function . If
the HPP
(exponential)
model as a
special case then we have an NHPP with a Power Law intensity function (the
"intensity function" is another name for the repair rate m(t)).
Because of the polynomial nature of the ROCOF, this model is very
flexible and can model both increasing (b>1 or < 0) and decreasing
(0 < b < 1 or 0 < < 1)) failure rates. When b = 1 or = 0, the model
reduces to the HPP constant repair rate model.

Probabilities Probabilities of a given number of failures for the NHPP model are
of failure for calculated by a straight forward generalization of the formulas for the
all NHPP HPP. Thus, for any NHPP
processes
can easily be
calculated
based on the
Poisson and for the Power Law model:
formula
The Power Other names for the Power Law model are: the Duane Model and the
Law model AMSAA model. AMSAA stands for the United States Army
is also called Materials System Analysis Activity, where much theoretical work
the Duane describing the Power Law model was done in the 1970's.
Model and
the AMSAA
model
It is also The time to the first fail for a Power Law process has a Weibull
called a distribution with shape parameter b and characteristic life a. For this
Weibull reason, the Power law model is sometimes called a Weibull Process.
Process - but This name can be confusing and should be avoided. However, since it
this name is mixes a life distribution model applicable to lifetimes of a non
misleading repairable population with a model for inter arrival times of failures of a
repairable population.
For any NHPP process with intensity function m(t), the distribution
function (CDF) for the interarrival time to the next failure, given a
failure just occurred at time T, is given by

Once a
failure
occurs, the
waiting time
to the next
In particular, for the Power Law the waiting time to the next failure,
failure for
given a failure at time T, has distribution function
an NHPP
has a simple
CDF
formula
This inter arrival time CDF can be used to derive a simple algorithm for
simulating NHPP Power Law Data.

8.1.7.3. Exponential law

8.1. Introduction
8.1.7.3. Exponential law

The An NHPP with ROCOF or intensity function given by
Exponential
Law is
another
flexible is said to follow an Exponential Law. This is also called the log linear
NHPP model model or the Cox-Lewis model.
A system whose repair rate follows this flexible model is improving if
< 0 and deteriorating if >0. When = 0, the Exponential Law
reduces to the HPP constant repair rate model

8.1.8. How can you evaluate reliability from the "bottom - up" (component failure mode to system failure rates)?

8.1. Introduction
8.1.8. How can you evaluate reliability from

the "bottom - up" (component failure
mode to system failure rates)?
Several This section deals with models and methods that apply to non
simple repairable components and systems. Models for failure rates (and not
models can repair rates) are described. The next section covers models for
be used to (repairable) system reliability growth.
calculate
system We use the Competing Risk Model to go from component failure
failure rates, modes to component failure rates. Next we use the Series Model to go
starting with from components to assemblies and systems. These models assume
failure rates independence and "first failure mode to reach failure causes both the
for failure component and the system to fail".
modes within
individual If some components are "in parallel", so that the system can survive one
system (or possibly more) component failures, we have the parallel or
components redundant model. If an assembly has n identical components, at least r
of which must be working for the system to work, we have what is
known as the r out of n model.
The standby model uses redundancy like the parallel model, except that
the redundant unit is in an off-state (not exercised) until called upon to
replace a failed unit.
This section describes these various models. The last subsection show
how complex systems can be evaluated using the various models as
building blocks.

8.1.8.1. Competing risk model

8.1. Introduction
8.1.8. How can you evaluate reliability from the "bottom - up" (component failure mode to system
failure rates)?

Use the Assume a (replaceable) component or unit has k different ways it can
competing fail. These are called failure modes and underlying each failure mode
risk model there is a failure mechanism.
when the
failure The Competing Risk Model evaluates component reliability by
mechanisms "building up" from the reliability models for each failure mode.
are The following 3 assumptions are all needed:
independent
1. Each failure mechanism leading to a particular type of failure (i.e.
and the first
failure mode) proceeds independently of every other one, at least
mechanism
until a failure occurs.
failure
causes the 2. The component fails when the first of all the competing failure
component mechanisms reaches a failure state.
to fail. 3. Each of the k failure modes has a known life distribution model
Fi(t).
The competing risk model can be used when all three assumptions hold.
If Rc(t), Fc(t), and hc(t) denote the reliability, CDF and failure rate for
the component and Ri(t), Fi(t) and hi(t) are the reliability, CDF and
failure rate for the i-th failure mode, then the competing risk model
formulas are:

Multiply
reliabilities
and add
failure rates
Think of the competing risk model in the following way:

All the failure mechanisms are having a race to see which
can reach failure first. They are not allowed to "look over
their shoulder or sideways" at the progress the other ones
are making. They just go their own way as fast as they can
and the first to reach "failure" causes the component to
fail.
Under these conditions the component reliability is the
product of the failure mode reliabilities and the component
failure rate is just the sum of the failure mode failure rates.
Note that the above holds for any arbitrary life distribution models, as
long as "independence" and "first mechanism failure causes the
component to fail" holds.
When we learn how to plot and analyze reliability data in later sections,
the methods will be applied separately to each failure mode within the
data set (considering failures due to all other modes as "censored run
times"). With this approach, the competing risk model provides the glue
to put the pieces back together again.

8.1.8.2. Series model

8.1. Introduction

The series The Series Model is used to build up from components to sub-assemblies and systems.
model is used It only applies to non replaceable populations (or first failures of populations of
to go from systems). The assumptions and formulas for the Series Model are identical to those for
individual the Competing Risk Model, with the k failure modes within a component replaced by the
components n components within a system.
to the entire
system The following 3 assumptions are all needed:
assuming the 1. Each component operates or fails independently of every other one, at least until
system fails the first component failure occurs.
when the first 2. The system fails when the first component failure occurs.
component 3. Each of the n (possibly different) components in the system has a known life
fails and all distribution model Fi(t).
components
fail or
survive
independently
of one
another
Add failure When the Series Model assumptions hold we have:

rates and
multiply
reliabilities
in the Series
Model
where the subscript S refers to the entire system and the subscript i refers to the i-th
component.
Note that the above holds for any arbitrary component life distribution models, as long
as "independence" and "first component failure causes the system to fail" both hold.
The analogy to a series circuit is useful. The entire system has n components in series.
The system fails when current no longer flows, and each component operates or fails
independently of all the others. The schematic below shows a system with 5 components
in series "replaced" by an "equivalent" (as far as reliability is concerned) system with

only one component.

8.1.8.3. Parallel or redundant model

8.1. Introduction

The parallel The opposite of a series model, where the first component failure causes the system to
model fail, is a parallel model where all the components have to fail before the system fails. If
assumes all n there are n components, any (n-1) of them may be considered redundant to the remaining
components one (even if the components are all different). When the system is turned on, all the
that make up components operate until they fail. The system reaches failure at the time of the last
a system component failure.
operate
independently The assumptions for a parallel model are:
and the 1. All components operate independently of one another, as far as reliability is
system works concerned.
as long as
2. The system operates as long as at least one component is still operating. System
one
failure occurs at the time of the last component failure.
component
still works. 3. The CDF for each component is known.
Multiply
component
CDF's to get
the system
CDF for a
parallel
model
For a parallel model, the CDF Fs(t) for the system is just the product of the CDF's Fi(t)
for the components or
Rs(t) and hs(t) can be evaluated using basic definitions, once we have Fs(t).
The schematic below represents a parallel system with 5 components and the (reliability)
equivalent 1 component system with a CDF Fs equal to the product of the 5 component
CDF's.


8.1.8.4. R out of N model

8.1. Introduction
failure rates)?

An r out of n An "r out of "n" system contains both the series system model and the
model is a parallel system model as special cases. The system has n components
system which operate or fail independently of one another and as long as at
which least r of these components (any r) survive, the system survives. System
survives failure occurs when the (n-r+1)th component failure occurs.
when at
least r of its When r = n, the r out of n model reduces to the series model. When r =
components 1, the r out of n model becomes the parallel model.
are working
We treat here the simple case where all the components are identical.
(any r).
Formulas and assumptions for r out of n model (identical components):
1. All components have the identical reliability function R(t).
2. All components operate independently of one another (as far as
failure is concerned).
3. The system can survive any (n-r) of the components failing. The
system fails at the instant of the (n-r+1)th component failure.
Formula for System reliability is given by adding the probability of exactly r

an r out of n components surviving to time t to the probability of exactly (r+1)
system components surviving, and so on up to the probability of all components
where the surviving to time t. These are binomial probabilities (with p = R(t)), so
components the system reliability is given by:
are identical
Note: If we relax the assumption that all the components are identical,
then Rs(t) would be the sum of probabilities evaluated for all possible
terms that could be formed by picking at least r survivors and the
corresponding failures. The probability for each term is evaluated as a

product of R(t)'s and F(t)'s. For example, for n = 4 and r = 2, the system
reliability would be (abbreviating the notation for R(t) and F(t) by using
only R and F)
Rs = R1R2F3F4 + R1R3F2F4 + R1R4F2F3 + R2R3F1F4
+ R2R4F1F3 + R3R4F1F2 + R1R2R3F4 + R1R3R4F2
+ R1R2R4F3 + R2R3R4F1 + R1R2R3R4

8.1.8.5. Standby model

8.1. Introduction
failure rates)?

The Standby A Standby Model refers to the case where a key component (or
Model assembly) has an identical backup component in an "off" state until
evaluates needed. When the original component fails, a switch turns on the
improved "standby" backup component and the system continues to operate.
reliability
when backup In the simple case, assume the non standby part of the system has CDF
replacements F(t) and there are (n-1) identical backup units which will operate in
are switched sequence until the last one fails. At that point, the system finally fails.
on when
failures The total system lifetime is the sum of n identically distributed random
occur. lifetimes, each having CDF F(t).
Identical In other words, Tn = t1 + t2+ ... + tn, where each ti has CDF F(t) and Tn
backup has a CDF we denote by Fn(t). This can be evaluated using convolution
Standby formulas:
model leads
to
convolution
formulas
In general, convolutions are solved numerically. However, for the

special case when F(t) is the exponential model, the above integrations
can be solved in closed form.

Exponential Special Case: The Exponential (or Gamma) Standby Model

standby
systems lead If F(t) has the exponential CDF (or F(t) = 1 - e-λt ), then
to a gamma
lifetime
model
and the PDF fn(t) is the well known gamma distribution.
Example: An unmanned space probe sent out to explore the solar

system has an onboard computer with reliability characterized by the
exponential distribution with a Mean Time To Failure (MTTF) of 1/
= 30 months (a constant failure rate of 1/30 = .033 fails per month). The
probability of surviving a two year mission is only e-24/30 = .45. If,
however, a second computer is included in the probe in a standby mode,
the reliability at 24 months (using the above formula for F2) becomes .8
× .449 + .449 = .81. The failure rate at 24 months (f2/[1-F2]) reduces to
[(24/900) ×.449]/.81 = .015 fails per month. At 12 months, the failure
rate is only .0095 fails per month, a reduction of more than 33 times as
compared to the non standby case.
Standby units (as the example shows) are an effective way of increasing
reliability and reducing failure rates, especially during the early stages
of product life. Their improvement effect is similar to, but greater than,
that of parallel redundancy . The drawback, from a practical standpoint,
is the expense of extra components that are not needed for
functionality.

8.1.8.6. Complex systems

8.1. Introduction

Often the Many complex systems can be diagrammed as combinations of Series components,
reliability Parallel components, R out of N components and Standby components. By using the
of complex formulas for these models, subsystems or sections of the original system can be replaced
systems can by an "equivalent" single component with a known CDF or Reliability function.
be Proceeding like this, it may be possible to eventually reduce the entire system to one
evaluated component with a known CDF.
by
successive Below is an example where a complex system composed of both components in parallel
applications and components in series is reduced first to a series system and finally to a one
of Series component system.
and/or
Parallel
model
formulas

Note: The reduction methods described above will work for many, but not all, systems.
Some systems with a complicated operational logic structure will need a more formal
structural analysis methodology. This methodology deals with subjects such as event
trees, boolean representations, coherent structures, cut sets and decompositions, and is
beyond the present scope of this Handbook.

8.1.9. How can you model reliability growth?

8.1. Introduction
8.1.9. How can you model reliability

growth?
A reliability During the early stages of developing and prototyping complex
improvement systems, reliability often does not meet customer requirements. A
test is a formal test procedure aimed at discovering and fixing causes of
formal unreliability is known as a Reliability Improvement Test. This test
procedure focuses on system design, assembly and component selection
aimed at weaknesses that cause failures.
discovering
and fixing A typical reliability improvement test procedure would be to run a
system prototype system as the customer might for a period of several weeks,
reliability while a multidisciplined team of engineers and technicians (design,
flaws quality, reliability, manufacturing, etc., ) analyze every failure that
occurs. This team comes up with root causes for the failures and
develops design and/or assembly improvements to hopefully eliminate
or reduce the future occurrence of that type of failure. As the testing
continues, the improvements the team comes up with are incorporated
into the prototype so it is expected that reliability will improve during
the course of the test.
Repair rates Another name for reliability improvement testing is TAAF testing,
should show standing for Test, Analyze And Fix. In the semiconductor industry,
an another common name for a reliability test (trademarked by Motorola)
improvement is an IRONMAN™. The acronym IRONMAN™ stands for "Improve
trend during Reliability Of New Machines At Night" and empathizes the "around the
the course of clock" nature of the testing process.
a reliability
improvement While only one model applies when a repairable system has no
test and this improvement or degradation trends (the constant repair rate HPP model)
can be there are infinitely many models that could be used to describe a system
modeled with a decreasing repair rate (reliability growth models).
using a
NHPP model Fortunately, one or two relatively simple models have been very
successful in a wide range of industrial applications. Two models that
have previously been described will be used in this section. These

models are the NHPP Power Law Model and the NHPP Exponential
Law Model. The Power Law Model underlies the frequently used
graphical technique known as Duane Plotting.

8.1.9.1. NHPP power law

8.1. Introduction

If the Power This repairable system model was described in Section 8.1.7.2. The expected number of
Law applies, failures by time t has the form M(t) = atb and the repair rate has the form m(t) = abtb-1 .
Repair Rates This will model improvement when 0<b<1, with larger improvements coming when b is
improve smaller. As we will see in the next section on Duane Plotting, it is convenient to define
over time
according to = 1 - b and = ab, and write the repair rate as
the formula
m(t) =
. The
exponent Again we have improvement when 0 < < 1, with larger improvement coming from
lies between larger values of . is known as the Duane Plot slope or the reliability improvement
0 and 1 and
is called the Growth Slope.
reliability In terms of the original parameters parameters for M(t), we have
growth
slope.
Use of the Power Law model for reliability growth test data generally assumes the
following:
1. While the test is on-going system improvements are introduced that produce continual
improvements in the rate of system repair.
2. Over a long enough period of time the effect of these improvements can be modeled
accurately by the continuous polynomial repair rate improvement model
When an
improvement 3. When the improvement test ends at test time T, and no further improvement actions are
test ends, the
MTBF stays on-going, the repair rate has been reduced to . The repair rate remains constant
constant at from then on at this new (improved) level.
its last
Assumption 3 means that when the test ends, the HPP constant repair rate model takes
achieved
over and the MTBF for the system from then on is the reciprocal of the final repair rate
value.
or . If we estimate the expected number of failures up to time T by the actual
number observed, the estimated MTBF at the end of a reliability test (following the
Power Law) is:

where T is the test time, ris the total number of test failures and is the reliability
growth slope. A formula for estimating from system failure times is given in the
Analysis Section for the Power Law model.
Simulated Simulating NHPP Power Law Data

Data
Example Step 1: User inputs the positive constants a and b.
Step 2: Simulate a vector of n uniform (0,1) random numbers. Call these U1, U2, U3, . . .
Un.
Step 3: Calculate Y1 = {-1/a * ln U1} ** 1/b
Step 2+i: Calculate Yi = {(Yi-1 ** b) -1/a * ln Ui} ** 1/b
Step 2+n: Calculate Yn = {(Yn-1 ** b) -1/a * ln Un} ** 1/b
The n numbers Y1, Y2, . . ., Ynare the desired repair times simulated from an NHPP Power
Law process with parameters a, b (or = 1 - b and = ab).
The Dataplot Macro powersim.dp will ask the user to input the number N of repair times
desired and the parameters a and b. The program will output the N simulated repair times
and a plot of these repair times.
Example
Below powersim.dp is used to generate 13 random repair times from the NHPP Power
Law process with a = .2 and b = .4.
CALL powersim.dp
Enter number N of simulated repair times desired
13
Enter value for shape parameter a (a > 0)
.2
Enter value for shape parameter b (b > 0)
.4
FAILNUM FAILTIME
1 26
2 182
3 321
4 728
5 896
6 1268
7 1507
8 2325
9 3427
10 11871
11 11978

12 13562
13 15053

8.1.9.2. Duane plots

8.1. Introduction

A plot on The standard estimate of the MTBF for a system with a constant repair rate (an HPP system) is
log-log T/r, where T is the total time the system was observed and r is the total number of failures that
paper of occurred.
successive
MTBF If we calculate successive MTBF estimates (called Cum MTBF Estimates) every time a failure
estimates occurs for a system undergoing reliability improvement testing, we typically see a sequence of
versus mostly increasing numbers.
system time In 1964, J. T. Duane observed that when he plotted these cum MTBF estimates versus the times
of fail for of failure on log-log paper, the points tended to line up following a straight line. This was true for
reliability many different sets of reliability improvement data and many other engineers have seen similar
improvement results over the last three decades. This type of plot is called a Duane Plot and the slope of the
test data is
called a best line through the points is called the reliability growth slope or Duane plot slope.
Duane Plot
Points on a Plotting a Duane Plot is simple. If the ith failure occurs at time ti, then plot ti divided by i (the
Duane plot "y"- axis value) versus the timeti (the "x"-axis value) on log-log graph paper. Do this for all the
line up test failures and draw the best straight line you can following all these points.
following a
straight line Why does this "work"? Following the notation for repairable system models, we are plotting
if the Power estimates of {t/M(t)} versus the time of failure t. If M(t) follows the Power Law (also described in
Law model the last section), then we are plotting estimates of t/atb versus the time of fail t. This is the same
applies
as plotting versus t, where = 1-b . On log-log paper this will be a straight line with
slope and intercept (when t = 1) of - log10a.
In other words, a straight line on a Duane plot is equivalent to the NHPP Power Law Model with
a reliability growth slope of = 1 - b and an "a" parameter equal to
10-intercept.
Note: A useful empirical rule of thumb based on Duane plots made from many reliability
improvement tests across many industries is the following:
Duane plot The reliability improvement slope for virtually all reliability improvement tests will
reliability lie between .3 and .6. The lower end (.3) describes a minimally effective test -
growth perhaps the cross functional team is inexperienced or the system has many failure
slopes mechanisms that are not well understood. The higher end (.6) approaches the
should lie empirical state of the art for reliability improvement activities.
between .3
and .6

Examples of Duane Plot Example 1:

Duane Plots
A reliability growth test lasted 1500 hours (approximately 10 weeks) and recorded 8 failures at
the following system hours; 33, 76, 145, 347, 555, 811, 1212, 1499. After calculating successive
cum MTBF estimates, a Duane plot shows these estimates versus system age on log vs log paper.
The "best' straight line through the data points corresponds to an NHPP Power Law model with
reliability growth slope equal to the slope of the line. This line is an estimate of the theoretical
model line (assuming the Power Law holds during the course of the test) and the achieved MTBF
at the end of test is given by T/[r×(1- )], where T is the total test time and r is the number of
failures. Results for this particular reliability growth test follow.
Failure # System Age of Failure Cum MTBF
1 33 33
2 76 38
3 145 48.3
4 347 86.8
5 555 111.0
6 811 135.2
7 1212 173.1
8 1499 187.3
The
Duane plot indicates a reasonable fit to a Power Law NHPP model. The reliability improvement
slope (slope of line on Duane plot) is = .437 (using the formula given in the section on
reliability data analysis for the Power Law model) and the estimated MTBF achieved by the end
of the 1500 hour test is 1500/(8 × [1-.437]) or 333 hours.
Duane Plot Example 2:
For the simulated Power Law data used in the Example in the preceding section, the following
Dataplot commands (executed immediately after running powersim.dp) produce the Duane Plot
shown below.
XLOG ON
YLOG ON
LET MCUM = FAILTIME/FAILNUM
PLOT MCUM FAILTIME


8.1.9.3. NHPP exponential law

8.1. Introduction
8.1.9.3. NHPP exponential law

The When the data points in a Duane plot show obvious curvature, another
Exponential model that might fit better is the NHPP Exponential Law.
Law is
another For this model, if < 0, the repair rate improves over time according
useful to
reliability
growth
model to try
when the The corresponding cumulative expected failures model is
Power law is
not fitting
well
This approaches the maximum value of A expected failures as t goes to
infinity, so the cumulative failures plot should clearly be bending over
and asymptotically approaching a value .
Rule of thumb: First try a Duane plot and the Power law model. If that
shows obvious lack of fit try the Exponential Law model, estimating
parameters using the formulas in the Analysis Section for the
Exponential law. A plot of cum fails versus time, along with the
estimated M(t) curve, can be used to judge goodness of fit.

8.1.10. How can Bayesian methodology be used for reliability evaluation?

8.1. Introduction
8.1.10. How can Bayesian methodology be used for

reliability evaluation?
Several This section gives an overview of the application of Bayesian techniques in reliability
Bayesian investigations. The following topics are covered:
Methods ● What is Bayesian Methodology ?
overview
topics are ● Bayes Formula, Prior and Posterior Distribution Models, and Conjugate Priors
covered in ● How Bayes Methodology is used in System Reliability Evaluation
this section ● Advantages and Disadvantages of using Bayes Methodology
What is Bayesian Methodology?
Bayesian It makes a great deal of practical sense to use all the information available, old or new,
analysis objective or subjective, when making decisions under uncertainty. This is especially true
considers when the consequences of the decisions can have a significant impact, financial or
population otherwise. Most of us make everyday personal decisions this way, using an intuitive process
parameters based on our experience and subjective judgments.
to be
random, not Main stream statistical analysis, however, seeks objectivity by generally restricting the
fixed. information used in an analysis to that obtained from a current set of clearly relevant data.
Prior knowledge is not used except to suggest the choice of a particular population model to
Old "fit" to the data, and this choice is later checked against the data for reasonableness.
information,
or subjective Lifetime or repair models, as we saw earlier when we looked at repairable and non
judgment, is repairable reliability population models, have one or more unknown parameters. The
used to come classical statistical approach considers these parameters as fixed but unknown constants to
up with a be estimated (i.e. "guessed at") using sample data taken randomly from the population of
prior interest. A confidence interval for an unknown parameter is really a frequency statement
distribution about the likelihood that numbers calculated from a sample capture the true parameter.
for these Strictly speaking, one cannot make probability statements about the true parameter since it
population is fixed, not random.
parameters.
The Bayesian approach, on the other hand, treats these population model parameters as
random, not fixed, quantities. Before looking at the current data, we use old information, or
even subjective judgments, to construct a prior distribution model for these parameters.
This model expresses our starting assessment about how likely various values of the
unknown parameters are. We then make use of the current data (via Bayes formula) to
revise this starting assessment, deriving what is called the posterior distribution model for
the population model parameters. Parameter estimates, along with confidence intervals
(known as credibility intervals), are calculated directly from the posterior distribution.
Credibility intervals are legitimate probability statements about the unknown parameters,
since these parameters now are considered random, not fixed.
It is unlikely in most applications that data will ever exist to validate a chosen prior
distribution model. Parametric Bayesian prior models are chosen because of their flexibility
http://www.itl.nist.gov/div898/handbook/apr/section1/apr1a.htm (1 of 3) [5/22/2001 1:28:31 PM]

and mathematical convenience. In particular, conjugate priors (defined below) are a natural
and popular choice of Bayesian prior distribution models.
Bayes Formula, Prior and Posterior Distribution Models, and Conjugate Priors
Bayes Bayes formula is a useful equation from probability theory that expresses the conditional
formula probability of an event A occurring, given that the event B has occurred (written P(A|B)), in
provides the terms of unconditional probabilities and the probability the event B has occurred, given that
mathematical A has occurred. In other words, Bayes formula inverts which of the events is the
tool that conditioning event. The formula is
combines
prior
knowledge
with current
data to and P(B) in the denominatorr is further expanded by using the so-called "Law of Total
produce a Probabilities" to write
posterior
distribution
where the events Ai are mutually exclusive and exhaust all possibilities and usually include
the event A as one of the Ai.
The same formula, written in terms of probability density function models, takes the form:
where f(x| ) is the probability model, or likelihood function, for the observed data x given
the unknown parameter (or parameters) , g( ) is the prior distribution model for and
g( |x) is the posterior distribution model for given that the data x have been observed.
When g( |x) and g( ) both belong to the same distribution family, then g( ) and
f(x| ) are called conjugate distributions and g( ) is the conjugate prior for f(x| ). For
example, the Beta distribution model is a conjugate prior for the proportion of successes p
when samples have a binomial distribution. And the Gamma model is a conjugate prior for
the failure rate when sampling failure times or repair times from an exponentially
distributed population. This latter conjugate pair (gamma, exponential) is used extensively
in Bayesian system reliability applications.
How Bayes Methodology is used in System Reliability Evaluation

Bayesian Models and assumptions for using Bayes methodology will be described in a later section.
system Here we compare the classical paradigm versus the Bayesian paradigm when system
reliability reliability follows the HPP or exponential model (i.e., the flat portion of the Bathtub Curve).
evaluation
assumes the Classical Paradigm For System Reliability Evaluation:
system ● The MTBF is one fixed unknown value - there is no “probability” associated with it.
MTBF is a ● Failure data from a test or observation period allows you to make inferences about the
random value of the true unknown MTBF
quantity
● No other data is used and no “judgment” - the procedure is objective and based solely
"chosen"
according to on the test data and the assumed HPP model
a prior Bayesian Paradigm For System Reliability Evaluation:
distribution ● The MTBF is a random quantity with a probability distribution
model ● The particular piece of equipment or system you are testing “chooses” an MTBF from
this distribution and you observe failure data that follows an HPP model with that
MTBF
● Prior to running the test, you already have some idea of what the MTBF probability
distribution looks like based on prior test data or an consensus engineering judgment
Advantages and Disadvantages of using Bayes Methodology
Pro's and While the primary motivation to use Bayesian reliability methods is typically a desire to
con's for save on test time and materials cost, there are other factors that should also be taken into
using account. The table below summarizes some of these "good news" and "bad news"
Bayesian considerations.
methods
Bayesian Paradigm: Advantages and Disadvantages
Pro's Con's
● Uses prior information - this "makes ● Prior information may not be

sense" accurate - generating misleading
● If the prior information is encouraging, conclusions
less new testing may be needed to ● Way of inputting prior information
confirm a desired MTBF at a given (choice of prior) may not be correct
confidence ● Customers may not accept validity of
● Confidence intervals are really intervals prior data or engineering judgements
for the (random) MTBF - sometimes ● There is no one "correct way" of
called "credibility intervals" inputting prior information and
different approaches can give
different results
● Results aren't objective and don't
stand by themselves

8.2. Assumptions/Prerequisites
This section describes how life distribution models and acceleration
models are typically chosen. Several graphical and analytical methods
for evaluating model fit are also discussed.
Detailed 2. Assumptions/Prerequisites
contents of 1. How do you choose an appropriate life distribution model?
Section 2
1. Based on failure mode
2. Extreme value argument
3. Multiplicative degradation argument
4. Fatigue life (Birnbaum-Saunders) argument
5. Empirical model fitting - distribution free (Kaplan-Meier)
approach
2. How do you plot reliability data?
1. Probability plotting
2. Hazard and cum hazard plotting
3. Trend and growth plotting (Duane plots)
3. How can you test reliability model assumptions?
1. Visual tests
2. Goodness of fit tests
3. Likelihood ratio tests
4. Trend tests
4. How do you choose an appropriate physical acceleration model?
5. What models and assumptions are typically made when Bayesian
methods are used for reliability evaluation


8.2.1. How do you choose an appropriate life distribution model?

8.2.1. How do you choose an appropriate

life distribution model?
Choose Life distribution models are chosen for one or more of the following
models that three reasons:
make sense, 1. There is a physical/statistical argument that theoretically matches
fit the data a failure mechanism to a life distribution model
and,
2. A particular model has previously been used successfully for the
hopefully,
same or a similar failure mechanism
have a
plausible 3. A convenient model provides a good empirical fit to all the failure
theoretical data
justification. Whatever method is used to choose a model, the model should
● "make sense" - for example, don't use an exponential model with
a constant failure rate to model a "wear out" failure mechanism
● pass visual and statistical tests for fitting the data.
Models like the lognormal and the Weibull are so flexible that it is not
uncommon that both will fit a small set of failure data equally well. Yet,
especially when projecting via acceleration models to a use condition far
removed from the test data, these two models may predict failure rates
that differ by orders of magnitude. That is why it is more than an
academic exercise to try to find a theoretical justification for using one
particular distribution over another.
There are We will consider three well known arguments of this type:
several ● Extreme value argument
useful
theoretical ● Multiplicative degradation argument
arguments ● Fatigue life (Birnbaum-Saunders) model
to help Note that physical/statistical arguments for choosing a life distribution
guide the model are typically based on individual failure modes.
choice of a
model

For some Kaplan-Meier technique can be used when it is appropriate to just "let
questions, the data points speak for themselves" without making any model
an assumptions. However, you generally need lots of data for this approach
"empirical" to be useful, and acceleration modeling is much more difficult.
distribution
free
approach
can be used

8.2.1.1. Based on failure mode

8.2.1.1. Based on failure mode

Life Failure mode data is failure data sorted by types of failures. Root cause
distribution analysis must be done on each failure incident in order to characterize
models and them by failure mode. While this may be difficult and costly, it is a key
physical part of any serious effort to understand, model, project and improve
acceleration component or system reliability.
models
typically The natural place to apply both life distribution models and physical
only apply acceleration models is at the failure mode level. Each failure mode a
at the component can fail by will typically have a its own life distribution
individual model. The same is true for acceleration models. For the most part, these
failure mode models only make sense at the failure mode level, and not at the
level component or system level. Once each mode (or mechanism) is
modeled, the bottoms-up approach can be used to build up to the entire
component or system.
In particular, the arguments for choosing life distribution models
described in the next 3 sections apply at the failure mode level only.
These are the Extreme value argument, the Multiplicative degradation
argument and the Fatigue life (Birnbaum-Saunders) model.
The distribution free (Kaplan - Meier) approach can be applied at any
level (mode, component, system, etc.).

8.2.1.2. Extreme value argument


If component It is well known that the Central Limit Theorem predicts that normal
or system distributions will successfully model most engineering data where the
failure observed measurements arise from the sum of many small random
occurs when sources (such as measurement errors or control chart sample means).
the first of Practical experience validates this theory - the normal distribution
many "works" for many engineering data sets.
competing
failure Less well known is the fact that Extreme Value Theory predicts that the
processes Weibull distribution will successfully model failure times for
reaches a mechanisms where many competing similar failure processes are racing
critical to failure and the first to reach it (i.e. the minimum of a large collection
point, then of roughly comparable random failure times) produces the observed
Extreme failure time. In other words, when a large number of roughly equivalent
Value runners are competing, and the winning time is recorded for many
Theory similar races, these times are likely to follow a Weibull distribution.
predicts the Note that this does not mean that anytime there are several failure
Weibull mechanisms competing to cause a component or system to fail, the
Distribution Weibull model applies. One or a few of these mechanisms may
will be a dominate over the others and cause almost all of the failures - then the
good model "minimum of a large number of roughly comparable" random failure
times does not apply and the proper model should be derived from the
distribution models for the few dominating mechanisms using the
competing risk model.
On the other hand, there are many cases where failure occurs at the
weakest link of a large number of similar degradation processes or
defect flaws. One example of this occurs when modeling catastrophic
failures of capacitors caused by dielectric material breakdown. Typical
dielectric material has many "flaws" or microscopic sites where a
breakdown will eventually take place. These sites may be thought of as
competing with each other to reach failure first. The Weibull model,
as extreme value theory would predict, has been very successful as a
life distribution model for this failure mechanism.


8.2.1.3. Multiplicative degradation Argument

8.2.1.3. Multiplicative degradation

Argument
The
lognormal A brief verbal description of the multiplicative degradation argument
model can be (leading to a derivation of the lognormal model) was given under Uses
derived when of the Lognormal Distribution Model. Here a formal derivation will be
degradation outlined because it gives insight into why the lognormal has been a
is caused by successful model for many failure mechanisms based on degradation
random processes.
shocks which
Let y1, y2, ...yi be measurements of the amount of degradation for a
increase
degradation particular failure process taken at successive discrete instants of time as
at a rate the process moves towards failure. Assume the following relationships
proportional exist between the y's:
to the total
amount
already
present
where the are small independent random perturbations or "shocks"
to the system that move the failure process along. In other words, the
increase in the amount of degradation from one instant to the next is a
small random multiple of the total amount of degradation already
present. This is what is meant by multiplicative degradation. The
situation is analogous to a snowball rolling down a snow covered hill;
the larger it gets, the faster it grows because it is able to pick up even
more snow.
We can express the total amount degradation at the n-th instant of time
by

8.2.1.3. Multiplicative degradation Argument
where x0 is a constant and the are small random shocks. Next we

take natural logarithms of both sides and obtain:
Using a Central Limit Theorem argument we can conclude that ln xn

has a normal distribution. But by the properties of the lognormal
distribution, that means that xn(or the amount of degradation) will
follow a lognormal model for any n (or at any time t). Since failure
occurs when the amount of degradation reaches a critical point, time of
failure will be modeled successfully by a lognormal for this type of
process.
Failure What kinds of failure mechanisms might be expected to follow a

mechanisms multiplicative degradation model? The processes listed below are likely
that might be candidates:
successfully 1. Chemical reactions leading to the formation of new compounds
modeled by
2. Diffusion or migration of ions
the
lognormal 3. Crack growth or propagation.
distribution Many semiconductor failure modes are caused by one of these three
based on the degradation processes. Therefore, it is no surprise that the lognormal
multiplicative model has been very successful for the following semiconductor wear
degradation out failure mechanisms:
model 1. Corrosion
2. Metal migration
3. Electromigration
4. Diffusion
5. Crack growth

8.2.1.4. Fatigue life (Birnbaum-Saunders) model


model
A model The derivation of the Fatigue Life model is based on repeated cycles
derived from of stress causing degradation leading to eventual failure. The typical
random crack example is crack growth. One key assumption is that the amount of
growth taking degradation during any cycle is independent of the degradation in
place during any other cycle, with the same random distribution.
many
independent When this assumption matches well with an hypothesized physical
cycles of stress model describing the degradation process, one would expect the
Birnbaum-Saunders model to be a reasonable distribution model
candidate. (See the note in the derivation for comments about the
difference between the lognormal model derivation and the Fatigue
Life model assumptions. Also see the comment on Miner's Rule).

8.2.1.5. Empirical model fitting - distribution free (Kaplan-Meier) approach

8.2.1.5. Empirical model fitting - distribution

free (Kaplan-Meier) approach
The Kaplan- The Kaplan-Meier (K-M) Product Limit procedure provides quick,
Meier simple estimates of the Reliability function or the CDF based on failure
procedure data that may even be multicensored. No underlying model (such as
gives CDF Weibull or lognormal) is assumed; K-M estimation is an empirical
estimates for (non-parametric) procedure. Exact times of failure are required,
complete or however.
censored
sample data Calculating Kaplan - Meier Estimates
without
The steps for calculating K-M estimates are the following:
assuming a
particular 1. Order the actual failure times from t1 through tr, where there are r
distribution failures
model 2. Corresponding to each ti, associate the number ni, where ni = the
number of operating units just before the the i-th failure occurred
at time ti
3. Estimate R(t1) by (n1-1)/n1
4. Estimate R(ti) by R(ti-1) × (ni-1)/ni
5. Estimate the CDF F(ti) by 1 - R(ti)
Note that unfailed units taken off test (i.e. censored) only count up to the
last actual failure time before they were removed. They are include in
the ni counts up to and including that failure time, but not after.

Example of A simple example will illustrate the K-M procedure. Assume 20 units
K-M are on life test and 6 failures occur at the following times: 10, 32, 56, 98,
estimate 122, and 181 hours. There were 4 unfailed units removed from test for
calculations other experiments at the following times: 50 100 125 and 150 hours.
The remaining 10 unfailed units were removed from test at 200 hours.
The K-M estimates for this life test are:
R(10) = 19/20
R(32) = 19/20 x 18/19
R(56) = 19/20 x 18/19 x 16/17
R(98) = 19/20 x 18/19 x 16/17 x 15/16
R(122) = 19/20 x 18/19 x 16/17 x 15/16 x 13/14
R(181) = 19/20 x 18/19 x 16/17 x 15/16 x 13/14 x 10/11
A General Expression for K-M Estimates
A general expression for the K-M estimates can be written. Assume we
have n units on test and order the observed times for these n units from
t1 to tn. Some of these are actual failure times and some are running
times for units taken off test before they fail. Keep track of all the
indices corresponding to actual failure times. Then the K-M estimates
are given by:
where the "hat" over R indicates it is an estimate and S is the set of all
subscripts j such that tj is an actual failure time. The notation j S and tj
less than or equal to ti means we only form products for indices j that are
in S and also correspond to times of failure less than or equal to ti.
Once values for R(ti) are calculated, the CDF estimates are
F(ti) = 1 - R(ti)

A small Modified K-M Estimates

modification
of K-M The K-M estimate at the time of the last failure is R(tr) = 0 and F(tr) =
estimates 1. This estimate has a pessimistic bias and cannot be plotted (without
produces modification) on probability paper, since the CDF for standard
better reliability models asymptotically approaches 1 as time approaches
results for infinity. Better estimates for graphical plotting can be obtained by
probability modifying the K-S estimates so that they reduce to the median rank
plotting estimates for plotting Type I Censored life test data (described in the
next section). Modified K-M estimates are given by the formula
Once values for R(ti) are calculated, the CDF estimates are F(ti) = 1 -
R(ti)

8.2.2. How do you plot reliability data?


Plot Graphical plots of reliability data are quick useful visual tests of
reliability whether using a particular model is consistent with the observed data.
data on the The basic idea behind virtually all graphical plotting techniques is the
right following:
"special"
graph paper Points calculated from the data are placed on specially
and if the constructed graph paper and, as long as they line up
points line up approximately on a straight line, the analyst can conclude
approximately that the data are consistent with the particular model the
on a straight paper is designed to test.
line, the If the reliability data consists of (possibly multicensored) failure data
assumed from a non repairable population (or a repairable population where
model is a only time to the first failure is considered) then the models are life
reasonable distribution models such as the exponential, Weibull or lognormal. If
fit. the data consists of repair times for a repairable system, then the model
might be the NHPP Power Law and the plot would be a Duane Plot.
The kinds of plots we will consider for failure data from non repairable
populations are:
● Probability (CDF) plots
● Hazard and Cum Hazard plots

For repairable populations we have
● Trend plots (to check whether an HPP or exponential model
applies)
● Duane plots (to check whether the NHPP Power Law applies)
Later on (Section 8.4.2.1) we will also look at plots that can be used to
check acceleration model assumptions.
Note: Many of the plots discussed in this section can also be used to
get quick estimates of model parameters. This will be covered in later
sections. While there may be other more accurate ways of estimating
parameters, simple graphical estimates can be very handy, especially

when other techniques require software programs that are not readily
available.

8.2.2.1. Probability plotting


Use Probability plots are simple visual ways of summarizing reliability data by plotting CDF
probability estimates vs time on specially constructed probability paper.
plots to see
your data Commercial papers are available for all the typical life distribution models. One axis (some
and visually papers use the y-axis and others the x-axis, so you have to check carefully) is labeled "Time"
check and the other axis is labeled "Cum Percent" or "Percentile". There are rules, independent of the
model model or type of paper, for calculating plotting positions from the reliability data. These only
assumptions depend on the type of censoring in the data, and whether exact times of fail are recorded or or
only readout times.
Plot each Remember that different failure modes can and should be separated out and individually
failure analyzed. When analyzing failure mode A, for example, treat failure times from failure modes
mode B, C, etc., as censored run times. Then repeat for failure mode B, and so on.
separately
Data points When the points are plotted the analyst fits a straight line through them (either by eye, or with
line up the aid of a least squares fitting program). Every straight line on, say, Weibull paper, uniquely
roughly on corresponds to a particular Weibull life distribution model and the same is true for lognormal
a straight or exponential paper. If the points follow the line reasonably well, then the model is consistent
line when with the data - if it was your previously chosen model there is no reason to question the
the model choice. Depending on the type of paper, there will be a simple way to find the parameter
chosen is estimates that correspond to the fitted straight line.
reasonable
Plotting Plotting Positions: Censored Data (Type I or Type II)

positions on
probability At the time ti of the i-th failure, we need an estimate of the CDF (or the Cum. Population
paper Percent Failure). The simplest and most obvious estimate is just 100 × i/n (where there are a
depend on total of n units on test). This, however, is generally an overestimate (i.e. biased). Various texts
the type of recommend corrections such as 100 × (i-.5)/n or 100 × i/(n+1). Here, we recommend what are
data known as (approximate) median rank estimates:
censoring
Corresponding to the time ti of the i-th failure use a CDF or Percentile estimate of 100 × (i -
.3)/(n + .7)
Plotting Positions: Readout Data
Let the readout times be T1, T2, ..., Tk and let the corresponding new failures recorded at each
readout be r1, r2, ..., rk. Again, there are a total of n units on test.
Corresponding to the readout time Tj, use a CDF or Percentile estimate of

Plotting Positions: Multicensored Data

The calculations are more complicated for multicensored data. K-M estimates (described in a
preceding section) can be used to get plotting positions at every failure time. The more
accurate Modified K-M Estimates are recommended. They reduce to the Type I or II median
rank estimates when the data consists of only failures, without any removals except possibly at
the end of the test.
How Special Papers Work
It is not The general idea is to take the model CDF equation and write it in such a way that a function
difficult to of F(t) is a linear equation of a function of t. This will be clear after a few examples. In the
do formulas that follow, "ln" always means "take natural logarithms, while "log" always means
probability "take base 10 logarithms".
plotting for
many a) Exponential Model: Take the exponential CDF and rewrite it as
reliability
models even
without
specially
constructed
graph
papers.
If we let y = 1/{1 - F(t)} and x = t, then log y is linear in x, with slope /ln10. This shows we
can make our own special exponential probability paper by using standard semi log paper
(with a logarithmicy-axis). Use the plotting position estimates for F(ti) described above
(without the 100 × multiplier) to calculate pairs of (xi,yi) points to plot.
If the data are consistent with an exponential model, the resulting plot will have points that
line up roughly on a straight line going through the origin with slope /ln10.
b) Weibull Model: Take the Weibull CDF and rewrite it as
If we let y = ln [1/{1-F(t)}] and x = t, then log y is linear in x with slope This shows we can
make our own Weibull probability paper by using log log paper. Use the usual plotting
position estimates for F(ti) (without the 100 × multiplier) to calculate pairs of (xi,yi) points to

plot.
If the data are consistent with a Weibull model, the resulting plot will have points that line up
roughly on a straight line with slope . This line will cross the log x-axis at time t = and the
log y axis (i.e., the intercept) at - log .
c) Lognormal Model: Take the lognormal cdf and rewrite it as
where is the inverse function for the standard normal distribution (taking a probability as
an argument and returning the corresponding "z" value).
If we let y = t and x = {F(t)}, then log y is linear in x with slope /ln10 and intercept
(when F(t) = .5) of log T50. We can make our own lognormal probability paper by using semi
log paper (with a logarithmic y-axis). Use the usual plotting position estimates for F(ti)
(without the 100 × multiplier) to calculate pairs of (xi,yi) points to plot.
If the data are consistent with a lognormal model, the resulting plot will have points that line
up roughly on a straight line with slope /ln10 and intercept T50 on the y-axis.
d) Extreme Value Distribution (Type I - for minimum): Take the extreme value distribution
CDF and rewrite it as
If we let y = -ln(1 - F(x)), then ln y is linear in x with slope 1/ and intercept -µ / . We can
use semi log paper (with a logarithmic y-axis) and plot y vs x. The points should follow a
straight line with a slope of 1/ ln10 and an intercept of - ln10. The ln 10 factors are needed
because commercial log paper uses base 10 logarithms.
DATAPLOT Example
A Dataplot Using the Dataplot commands to generate Weibull random failure times, we generate 20
Weibull Weibull failure times with a shape parameter of γ = 1.5 and α = 500. Assuming a test time of
example of T = 500 hours, only 10 of these failure times would have been observed. They are, to the
probability nearest hour: 54, 187, 216, 240, 244, 335, 361, 373, 375, and 386. First we will compute
plotting plotting position CDF estimates based on these failure times, and then a probability plot using
the "make our own paper" method.
( 2) (3) (4)
( 1) F(ti) estimate ln{1/(1-F(ti)}
Time of Fail
Fail # = i
(x) (i-.3)/20.4 (y)
1 55 .034 .035
2 187 .083 .087
3 216 .132 .142

4 240 .181 .200

5 244 .230 .262
6 335 .279 .328
7 361 .328 .398
8 373 .377 .474
9 375 .426 .556
10 386 .475 .645
Of course, with commercial Weibull paper we would plot pairs of points from column (2) and
column (3). With ordinary log log paper we plot (2) vs (4).
The Dataplot sequence of commands and resulting plot follow:
LET X = DATA 55 187 216 240 244 335 361 373 375 386
LET Y = DATA .035 .087 .142 .2 .262 .328 .398 .474 .556 .645
XLOG ON
YLOG ON
XLABEL LOG TIME
YLABEL LOG LN (1/(1-F))
PLOT Y X
Note that the points appear to have some curvature. This is mostly due to the very first point
on the plot (the earliest time of failure). The first few points on a probability plot have more
variability than points in the central range and less attention should paid to them when visually
testing for "straightness".

Use of least We could go on to use Dataplot to fit a straight line through the points via the commands
squares
(regression) LET YY = LOG10(Y)
technique to LET XX = LOG10(X)
fit a line FIT YY XX
through the This would give a slope estimate of 1.46, which is close to the 1.5 value used in the
points on simulation.
probability The intercept is -4.114 and setting this equal to - log we estimate = 657 (the "true"
paper
value used in the simulation was 500).
Dataplot Finally, we note that Dataplot has a built in Weibull probability paper command that can be
has a used whenever we have a complete sample (i.e. no censoring and exact failure times). First
special you have to run PPCC to get an estimate of γ = GAMMA. This is stored under SHAPE. The
Weibull full sequence of commands (with XLOG and YLOG both set to OFF) is
probability
paper SET MINMAX = 1
function for WEIBULL PPCC PLOT SAMPLE
complete SET GAMMA = SHAPE
data WEIBULL PLOT SAMPLE
samples (no
censoring)

8.2.2.2. Hazard and cum hazard plotting


Another Just as most life distribution models have commercial probability paper for probability
kind of plotting reliability data, there are also special Cum Hazard Plotting papers available for
plotting, many life distribution models. These papers plot estimates for the Cum Hazard H(ti) vs
called the time ti of the i-th failure. As with probability plots, the plotting positions are
Cum
calculated independently of the model or paper used and a reasonable straight line fit to
Hazard
the points confirms that the chosen model and the data are consistent.
Plotting,
has the Advantages of Cum Hazard Plotting
same 1. It is much easier to calculate plotting positions for multicensored data using cum
purpose as hazard plotting techniques.
probability
plotting 2. Linear graph paper can be used for exponential data and log log paper can be used
for Weibull data.
Disadvantages of Cum Hazard Plotting
1. Commercial Cum Hazard paper may be difficult to find.
2. It is less intuitively clear just what is being plotted - cum percent failed (i.e.
probability plots) is meaningful and the resulting straight line fit can be used to
read off times when desired percents of the population will have failed. Percent
cumulative hazard increases beyond 100% and is harder to interpret.
3. Normal cum hazard plotting techniques require exact times of failure and running
times.
4. With computers to calculate the K-M estimates for probability plotting, the main
advantage of cum hazard plotting goes away.
Since probability plots are generally more useful, we will only give a brief description of
hazard plotting.
How to Make Cum Hazard Plots
1. Order the failure times and running times for each of the n units on test in
ascending order from 1 to n. The order is called the rank of the unit. Calculate the
reverse rank for each unit (reverse rank = n- rank +1).
2. Calculate a Hazard "value" for every failed unit (do this only for the failed units).
The Hazard value for the failed unit with reverse rank k is just 1/k.
3. Calculate the cumulative hazard values for each failed unit. The cumulative hazard
value corresponding to a particular failed unit is the sum of all the hazard values
for failed units with ranks up to and including that failed unit.
4. Plot the time of fail vs the cumulative hazard value on special Cum Hazard paper
(or construct your own paper as covered below for the exponential or the Weibull
model).

A life test Example: Ten units were tested at high stress test for up to 250 hours. Six failures took
cum place at 37, 73, 132, 195, 222 and 248 hours. Four units were taken off test without
hazard failing at the following run times: 50, 100, 200 and 250 hours. Cum hazard values were
plotting computed in the following table:
example
(2) (5)
(1) (3) (4) (6)
1= failure Haz Val
Time of Event Rank Reverse Rank Cum Hazard Value
0=runtime (2) x 1/(4)
37 1 1 10 1/10 .10
50 0 2 9
73 1 3 8 1/8 .225
100 0 4 7
132 1 5 6 1/6 .391
195 1 6 5 1/5 .591
200 0 7 4
222 1 8 3 1/3 .924
248 1 9 2 1/2 1.424
250 0 10 1
Next ignore the rows with no cum hazard value and plot column (1) vs column (6).
As with Exponential and Weibull "Homemade" Hazard Paper

probability
plotting, The cum hazard for the exponential is just H(t) = t, which is linear in t with 0
you can intercept. So a simple linear graph paper plot of y = col (6) vs x = col (1) should line up
make your
own as a straight line going through the origin with slope , if the exponential model is
special appropriate. The Dataplot commands and graph of this is shown below:
hazard LET X = DATA 37 73 132 195 222 248
plotting LET Y = DATA .1 .225 .391 .591 .924 1.424
paper for PLOT Y X
many
models

The cum Hazard for the Weibull is , so a plot of y vs x on log log paper
should line up along a straight line with slope , if the Weibull model is appropriate.
The Dataplot commands and graph of this is shown below:
XLOG ON
YLOG ON
PLOT Y X

The equation of the least squares line fit through these points can be found from
LET YY = LOG10(Y)
LET XX = LOG10(X)
FIT Y X
The Weibull fit looks better, although the slope estimate is 1.27, which is not far from an
exponential model slope of 1. Of course, with a sample of just 10, and only 6 failures, it
is difficult to pick a model from the data alone.

8.2.2.3. Trend and growth plotting (Duane plots)

8.2.2.3. Trend and growth plotting (Duane

plots)
Repair rates Models for repairable systems were described earlier. These models
are typically look at how the cumulative number of fails (or the repair rate) changes
either nearly over time. The two models used with most success throughout industry
constant over are the HPP (constant repair rate or "exponential" system model) and
time or have the NHPP Power Law process (the repair rate is the polynomial m(t) =
good or bad
trends ).
Before doing a Duane Plot, there are a few simple trend plots that often
convey strong evidence of the presence or absence of a trend in the
repair rate over time. If there is no trend, an HPP model is reasonable.
If there is an apparent improvement or degradation trend, a Duane Plot
will provide a visual check for whether the NHPP Power law model is
consistent with the data.
A few simple These simple visual graphical tests for trends are
plots can 1. Plot cumulative failures vs system age (a step function that goes
help us up every time there is a new failure). If this plot looks linear,
decide there is no obvious improvement (or degradation) trend. A
whether bending downward indicated improvement; bending upward
trends are indicates degradation.
present
2. Plot the inter arrival times between new failures (in other word,
the waiting times between failures, with the time to the first
failure used as the first "inter arrival" time). If these tend up there
is improvement, if they trend down there is degradation.
3. Plot the reciprocals of the inter arrival times. Each reciprocal is a
new failure rate estimate based only on the waiting time since the
last fail). If these trend down, there is improvement; an upward
trend indicates degradation.

Trend plots Case Study 1: Reliability Improvement Test Data Use of Trend
and a Duane Plots and Duane Plots
Plot for
actual A prototype of a new, complex piece of equipment went through a
Reliability 1500 operational hours Reliability Improvement Test. During the test
Improvement there were 10 fails. As a part of the improvement process, a cross
Test data functional Failure Review Board made sure every failure was analyzed
down to root cause and design and parts selection fixes were
implemented on the prototype. The observed failure times were: 5, 40,
43, 175, 389, 712, 747, 795, 1299 and 1478 hours, with the test ending
at 1500 hours. The reliability engineer on the Failure Review Board
first made trend plots as described above, then made a Duane plot.
These plots (using EXCEL) follow.

Time Cum MTBF

5 5
40 20
43 14.3
175 43.75

389 77.8
712 118.67
747 106.7
795 99.4
1299 144.3
1478 147.8
Comments: The three trend plots all show an improvement trend. The
reason it might help to try all three is that there are examples where
trends show up clearer on one of these plots, as compared to another.
Formal statistical tests on the significance of this visual evidence of a
trend will be shown in the section on Trend Tests.
The points on the Duane Plot line up roughly on a straight line,

indicating the NHPP Power Law model is consistent with the data.
Estimates for the reliability growth slope and the MTBF at the end of
this test for this case study will be given in a later section.

8.2.3. How can you test reliability model assumptions?

8.2.3. How can you test reliability model

assumptions?
Models are Since reliability models are often used to project (extrapolate) failure
frequently rates or MTBF's that are well beyond the range of the reliability data
necessary - used to fit these models, it is very important to "test" whether the
but should models chosen are consistent with whatever data is available. This
always be section describes several ways of deciding whether models are
checked acceptable. These are:
1. Visual Tests
2. Goodness of Fit Tests
3. Likelihood Ratio Tests
4. Trend Tests

8.2.3.1. Visual tests


A visual test We have already seen many examples of visual tests of models. These
of a model is were: Probability Plots, Cum hazard Plots, Duane Plots and Trend Plots.
a simple In all but the Trend Plots, the model was "tested' by how well the data
plot that points followed a straight line. In the case of the Trend Plots, we looked
tells us at a for curvature away from a straight line (cum repair plots) or increasing
glance or decreasing size trends (inter arrival times and reciprocal inter arrival
whether the times).
model is
consistent These simple plots are a powerful diagnostic tool, since the human eye
with the can often detect patterns or anomalies in the data by studying graphs.
data That kind of invaluable information would be lost if the analyst only
used quantitative statistical test to check model fit. Every analysis
should include as many visual tests as are applicable.
Advantages of Visual Tests
1. Easy to understand and explain.
2. Can occasionally reveal patterns or anomalies in the data.
3. When a model "passes" a visual sanity test, it is unlikely any
quantitative statistical test will "reject" it (the human eye is less
forgiving and more likely to detect spurious trends when there are
just random chance patterns)

Combine Disadvantages of Visual Tests

visual tests 1. Visual tests are subjective.
with formal
2. They do not quantify how well or poorly a model fits the data.
quantitative
tests for the 3. They are little help in choosing between two or more competing
"best of both models that both appear to fit the data.
worlds" 4. Simulation studies show that correct models may often appear to
approach not fit well by sheer chance - it is hard to know when visual
evidence is strong enough to reject what was previously believed
a correct model.
You can retain the advantages of visual tests and remove the
disadvantages by combining data plots with formal statistical test of
goodness of fit or trend.

8.2.3.2. Goodness of fit tests

8.2.3.2. Goodness of fit tests

A Goodness General tests for checking the hypothesis that your data is consistent
of Fit test with a particular model are discussed in Chapter 7. Details and examples
checks on of the Chi Square Goodness of Fit test and the Kolmolgorov-Smirnov
whether (K-S) test are given in Chapter 1. The Chi Square test can be used with
your data is Type I or Type II censored data and readout data, if there are enough
reasonable
failures and readout times. The K-S test generally requires complete
or highly
samples, which limits its usefulness in reliability analysis.
unlikely,
given an These tests control the probability of rejecting a valid model as follows:
assumed ● the analyst chooses a confidence level designated by 100 × (1 -
distribution
model ).
● a test statistic is calculated from the data and compared to likely
values for this statistic assuming the model is correct.
● if the test statistic has a very unlikely value, or less than or equal
to an chance of occurring, where is a small value like .1 or

.05 or even .01, then the model is rejected.
So the risk of rejecting the right model is kept to or less, and the
choice of usually takes into account the potential loss or difficulties
incurred if the model is rejected.

8.2.3.3. Likelihood ratio tests


Likelihood Likelihood functions for reliability data are described in Section 4. Two
Ratio Tests ways we use likelihood functions to choose models or verify/validate
are a assumptions are:
powerful,
very general 1. Calculate the maximum likelihood of the sample data based on an
method of assumed distribution model (the maximum occurs when unknown
testing model parameters are replaced by their maximum likelihood estimates).
assumptions. Repeat this calculation for other candidate distribution models that also
However, appear to fit the data (based on probability plots). If all the models have
they require the same number of unknown parameters, and there is no convincing
special reason to choose one particular model over another based on the failure
software, not mechanism or previous successful analyses, then pick the model with
always the largest likelihood value.
readily
2. Many model assumptions can be viewed as putting restrictions on the
available.
parameters in a likelihood expression that effectively reduce the total
number of unknown parameters. Some common examples are:
Examples i) It is suspected that a data set, typically modeled by a

where Weibull distribution, actually comes from an exponential
assumptions model. The exponential distribution is a special case of the
can be tested Weibull, where the shape parameter has been set to 1. If
by the we write the Weibull likelihood function for the data, the
Likelihood exponential model likelihood function is obtained by
Ratio Test setting to 1, and the number of unknown parameters has
been reduced from two to one.
ii) Assume we have n cells of data from an acceleration
test, where each cell had a different operating temperature.
We assume a lognormal population model applies in every
cell. Without an acceleration model assumption, the
likelihood of the experimental data would be the product
of the likelihoods from each cell and there would be 2n

unknown parameters (a different T50 and for each cell).

If we assume an Arrhenius model applies, the total number
of parameters drops from 2n to just 3, the single common
and the Arrhenius A and H parameters. This
acceleration assumption "saves" (2n-3) parameters.
iii) We life test samples of product from two vendors. The
product is known to have a failure mechanism modeled by
the Weibull distribution, and we want to know whether
there is a difference in reliability between the vendors. The
unrestricted likelihood of the data is the product of the two
likelihoods, with 4 unknown parameters (the shape and
characteristic life for each vendor population). If, however,
we assume no difference between vendors, the likelihood
reduces to having only two unknown parameters (the
common shape and the common characteristic life). Two
parameters are "lost" by the assumption of "no difference".
Clearly, we could come up with many more examples like these three,
where an important assumption can be restated as a reduction or
restriction on the number of parameters used to formulate the likelihood
function of the data. In all these cases, there is a simple and very useful
way to test whether the assumption is consistent with the data.
The Likelihood Ratio Test Procedure
Details of Let L1 be the maximum value of the likelihood of the data without the
the additional assumption. In other words, L1 is the likelihood of the data
Likelihood with all the parameters unrestricted and maximum likelihood estimates
Ratio Test substituted for these parameters.
procedure
Let L0 be the maximum value of the likelihood when the parameters are
In general,
restricted (and reduced in number) based on the assumption. Assume k
calculations
parameters were lost (i.e. L0 has k less parameters than L1)..
are difficult
and need to Form the ratio = L0/L1. This ratio is always between 0 and 1 and the
be built into
the software less likely the assumption is, the smaller will be. This can be
you use quantified at a given confidence level as follows:
1. Calculate = -2 ln . The smaller is, the larger will be.
2. We can tell when is significantly large by comparing it to the

upper 100 × (1- ) percentile point of a Chi Square distribution

with k degrees of freedom. has an approximate Chi Square

distribution with k degrees of freedom and the approximation is
usually good, even for small sample sizes.
3. The likelihood ratio tests computes and rejects the
assumption if is larger than a Chi Square with k degrees of

freedom percentile, where the percentile corresponds to the
confidence level chosen by the analyst.
Note: While Likelihood Ratio test procedures are very useful and
widely applicable, they are difficult to calculate by hand, especially for
censored data, and appropriate software is necessary.

8.2.3.4. Trend tests


Formal In this section we look at formal statistical tests that can allow us to
Trend Tests quantitatively determine whether or not the repair times of a system
should show a significant trend (which may be an improvement or a
accompany degradation trend). The section on trend and growth plotting discussed
Trend Plots visual tests for trends - this section complements those visual tests by
and Duane adding several numerical tests.
Plots. Three
are given in Three statistical test procedures will be described:
this section 1. The Reverse Arrangement Test (a simple and useful test that has
the advantage of making no assumptions about a model for the
possible trend)
2. The Military Handbook Test (optimal for distinguishing between
"no trend' and a trend following the NHPP Power Law or Duane
model)
3. The Laplace Test (optimal for distinguishing between "no trend'
and a trend following the NHPP Exponential Law model)
The Reverse The Reverse Arrangement Test

Arrangement
Test (RAT Assume there are r repairs during the observation period and they
test) is simple occurred at system ages T1, T2, T3, ...Tr (we set the start of the
and makes no observation period to T = 0). Let I1 = T1,
assumptions I2 = T2 - T1, I3 = T3 - T2, ..., Ir = Tr - Tr-1 be the interarrival times for
about what repairs (i.e. the sequence of waiting times between failures). Assume
model a trend the observation period ends at time Tend>Tr.
might follow
Previously, we plotted this sequence of inter arrival times to look for
evidence of trends. Now, we calculate how many instances we have of
a later inter arrival time being strictly greater than an earlier inter
arrival time. Each time that happens, we call it a reversal. If there are a
lot of reversals (more than likely from pure chance with no trend) we
have significant evidence of an improvement trend. If there are too few

reversals we have significant evidence of degradation.

A formal definition of the reversal count and some properties of this
count are:
● count a reversal every time Ij < Ik for some j and k with j < k
●this reversal count is the total number of reversals R

● for r repair times, the maximum possible number of reversals is
r(r-1)/2
● if there are no trends, on the average one would expect to have
r(r-1)/4 reversals.
As a simple example, assume we have 5 repair times at system ages 22,
58, 71, 156 and 225 and the observation period ended at system age
300 . First calculate the inter arrival times and obtain: 22, 36, 13, 85,
69. Next count reversals by "putting your finger" on the first inter
arrival time, 22, and counting how many later inter arrival times are
greater than that. In this case, there are 3. Continue by "moving your
finger to the second time, 36, and counting how many later times are
greater. There are exactly 2. Repeating this for the third and fourth
inter arrival times (with many repairs, your finger gets very tired!) we
obtain 2 and 0 reversals. Adding 3 + 2 + 2 + 0 = 7, we see that R = 7.
The total possible number of reversals is 5x4/2 = 10 and an "average
number is half this or 5.
In the example, we saw 7 reversals (2 more than average). Is this
strong evidence for an improvement trend? The following table allows
us to answer that at a 90% or 95% or 99% confidence level - the higher
the confidence, the stronger the evidence of improvement (or the less
likely that pure chance alone produced the result).
A useful table Value of R Indicating Significant Improvement (One-Sided Test)

to check Minimum R for Minimum R for Minimum R for
whether a Number of
90% Evidence of 95% Evidence of 99% Evidence of
reliability test Repairs
Improvement Improvement Improvement
has
4 6 6 -
demonstrated
significant 5 9 9 10
improvement 6 12 13 14
7 16 17 19
8 20 22 24
9 25 27 30
10 31 33 36
11 37 39 43
12 43 46 50

One-sided test means before looking at the data we expected

improvement trends, or, at worst, a constant repair rate. This would be
the case if we know of actions taken to improve reliability (such as
occur during reliability improvement tests).
For the r = 5 repair times example above where we had R = 7, the table
shows we do not (yet) have enough evidence to demonstrate a
significant improvement trend. That does not mean that an
improvement model is incorrect - it just means it is not yet "proved"
statistically. With small numbers of repairs, it is not easy to obtain
significant results.
For numbers of repairs beyond 12, there is a good approximation
formula that can be used to determine whether R is large enough to be
significant. Calculate
Use this
formula when
there are
more than 12
repairs in the
data set
and if z > 1.282, we have at least 90% significance. If z > 1.645, we
have 95% significance and a z > 2.33 indicates 99% significance. Since
z has an approximate standard normal distribution, the Dataplot
command
LET PERCENTILE = 100* NORCDF(z)
will return the percentile corresponding to z.
That covers the (one-sided) test for significant improvement trends. If,
on the other hand, we believe there may be a degradation trend (the
system is wearing out or being over stressed, for example) and we want
to know if the data confirms this, then we expect a low value for R and
we need a table to determine when the value is low enough to be
significant. The table below gives these critical values for R.
Value of R Indicating Significant Degradation Trend (One-Sided Test)
Maximum R for Maximum R for Maximum R for
Number of
90% Evidence of 95% Evidence of 99% Evidence of
Repairs
Degradation Degradation Degradation
4 0 0 -
5 1 1 0
6 3 2 1

7 5 4 2
8 8 6 4
9 11 9 6
10 14 12 9
11 18 16 12
12 23 20 16
For numbers of repairs r >12, use the approximation formula above,

with R replaced by [r(r-1)/2 - R].
Because of The Military Handbook Test

the success
the Duane This test is better at finding significance when the choice is between no
model has trend and a NHPP Power Law (Duane) model. In other words, if the
enjoyed with data comes from a system following the Power Law, this test will
industrial generally do better than any other when it comes to finding
improvement significance.
test data, this As before, we have r times of repair T1, T2, T3, ...Tr with the
Trend Test is observation period ending at time Tend>Tr. Calculate
recommended
and compare this to percentiles of the chi square distribution with 2r

degrees of freedom. For a one-sided improvement test, reject no trend
(or HPP) in favor of an improvement trend if the chi square value is
beyond an upper 90 (or 95, or 99) percentile. For a one-sided
degradation test, reject no trend if the chi square value is less than a 10
(or 5, or 1) percentile.
Trying out this test on the 5 repair times example, the test statistic has
value 13.28 with 10 degrees of freedom, and the following Dataplot
command evaluates the chi square percentile to be 79%:
LET PERCENTILE = 100*CHSCDF(13.28,10)
The Laplace Test
This test is better at finding significance when the choice is between no
trend and a NHPP Exponential model. In other words, if the data
comes from a system following the Exponential Law, this test will
generally do better than any other when it comes to finding
significance.

As before, we have r times of repair T1, T2, T3, ...Tr with the
observation period ending at time Tend>Tr. Calculate
and compare this to high (for improvement) or low (for degradation)

percentiles of the standard normal distribution. The Dataplot command
LET PERCENTILE = 100* NORCDF(z)
will return the percentile corresponding to z.
Formal tests Case Study 1: Reliability Test Improvement Data (Continued from
generally earlier)
confirm the
subjective The failure data and Trend plots and Duane plot were shown earlier.
information The observed failure times were: 5, 40, 43, 175, 389, 712, 747, 795,
conveyed by 1299 and 1478 hours, with the test ending at 1500 hours.
trend plots Reverse Arrangement Test: The inter arrival times are: 5, 35, 3, 132,
214, 323, 35, 48, 504 and 179. The number of reversals is 33, which,
according to the table above, is just significant at the 95% level.
The Military Handbook Test: The Chi Square test statistic using the
formula given above is 37.23 with 20 degrees of freedom. The
Dataplot expression
LET PERCENTILE = 100*CHSCDF(37.23,20)
yields a significance level of 98.9%. Since the Duane Plot looked very
reasonable, this test probably gives the most accurate significance
assessment of how unlikely it is that sheer chance produced such an
apparent improvement trend (only about 1.1% probability).

8.2.4. How do you choose an appropriate physical acceleration model?

8.2.4. How do you choose an appropriate

physical acceleration model?
Choosing a Choosing a physical acceleration model is a lot like choosing a life
good distribution model. First identify the failure mode and what stresses are
acceleration relevant (i.e. will accelerate the failure mechanism). Then check to see if
model is the literature contains examples of successful applications of a particular
part science model for this mechanism.
and part art
- but start If the literature offers little help try the models described in earlier
with a good sections :
literature ● Arrhenius
search ● The (inverse) pwer rule for voltage
● The exponential voltage model
● Two temperature/voltage models
● The electromigration model
● Three stress models (temperature, voltage and humidity)
● Eyring (for more than three stresses or when the above models
are not satisfactory)
● The Coffin-Manson mechanical crack growth model
All but the last model (the Coffin-Manson) apply to chemical or
electronic failure mechanisms and, since temperature is almost always a
relevant stress for these mechanisms, the Arrhenius model is nearly
always a part of any more general model. The Coffin-Manson works
well for many mechanical fatigue related mechanisms.
Sometimes models have to be adjusted to include a threshold level for
some stresses. In other words, failure might never occur due to a
particular mechanism unless a particular stress (temperature, for
example) is beyond a threshold value. A model for a temperature
dependent mechanism with a threshold at T = T0 might look like
time to fail = f(T)/(T-T0)

8.2.4. How do you choose an appropriate physical acceleration model?
where f(T) could be Arrhenius. As the temperature goes down towards

T0, time to fail increases towards infinity in this (deterministic)
acceleration model.
Models In some cases, a mathematical/physical description of the failure

derived mechanism can lead to an acceleration model. Some of the models
theoretically above were originally derived that way.
have been
very
successful
and are
convincing
Simple In general, use the simplest model (fewest parameters) you can. When
models are you have chosen a model, use visual tests and formal statistical fit tests
often the to confirm the model is consistent with your data. Continue to use the
best model as long as it gives results that "work," but be quick to look for a
new model when it is clear the old one is no longer adequate.
There are some good quotes that apply here:
Quotes from "All models are wrong, but some are useful." - George Box, and the
experts on principle of Occam's Razor (attributed to the 14th century logician
models William of Occam who said “Entities should not be multiplied
unnecessarily” - or something equivalent to that in Latin).
A modern version of Occam's Razor is: If you have two theories which
both explain the observed facts then you should use the simplest until
more evidence comes along - also called the Law of Parsimony.
Finally, for those who feel the above quotes place too much emphasis on
simplicity, there are several appropriate quotes from Albert Einstein:
"Make your theory as simple as possible, but no simpler"
"For every complex question there is a simple and wrong
solution."

8.2.5. What models and assumptions are typically made when Bayesian methods are usedfor reliability evaluation?

8.2.5. What models and assumptions are

typically made when Bayesian methods are
usedfor reliability evaluation?
The basics of Bayesian methodology were explained earlier, along with some of
the advantages and disadvantages of using this approach. Here we only consider
the models and assumptions that are commonplace when applying Bayesian
methodology to evaluate system reliability.
Bayesian Assumptions:
assumptions
for the 1. Failure times for the system under investigation can be adequately modeled
gamma by the exponential distribution. For repairable systems, this means the HPP
exponential model applies and the system is operating in the flat portion of the bathtub
system curve. While Bayesian methodology can also be applied to non repairable
model component populations, we will restrict ourselves to the system application in
this Handbook.
2. The MTBF for the system can be thought of as chosen from a prior
distribution model which is an analytic representation of our previous
information or judgments about the system's reliability. The form of this prior
model is the gamma distribution (the conjugate prior for the exponential
model). The prior model is actually defined for = 1/MTBF, since it is easier
to do the calculations this way.
3. Our prior knowledge is used to choose the gamma parameters a and b for the
prior distribution model for . There are many possible ways to convert
"knowledge" to gamma parameters, depending on the form of the "knowledge"
- we will describe three approaches.

Several i) If you have actual data from previous testing done on the system (or a
ways to system believed to have the same reliability as the one under
choose the investigation) this is the most credible prior knowledge, and the easiest to
prior use. Simply set the gamma parameter a equal to the total number of
gamma failuress from all the previous data, and set the parameter b equal to the
parameter total of all the previous test hours.
values
ii) A consensus method for determining a and b that works well is the
following: Assemble a group of engineers who know the system and its
sub components well from a reliability viewpoint.
❍ Have the group reach agreement on a reasonable MTBF they

expect the system to have. They could each pick a number they
would be willing to bet even money that the system would either
meet or miss, and the average or median of these numbers would
be their 50% best guess for the MTBF. Or they could just discuss
even money MTBF candidates until a consensus is reached
❍ Repeat the process again, this time reaching agreement on a low

MTBF they expect the system to exceed. A "5%" value that they
are "95% confident" the system will exceed (i.e., they would give
19 to 1 odds) is a good choice. Or a "10%" value might be chosen
(i.e., they would give 9 to 1 odds the actual MTBF exceeds the low
MTBF). Use whichever percentile choice the group prefers.
❍ Call the reasonable MTBF MTBF50 and the low MTBF you are
95% confident the system will exceed MTBF05. These two
numbers uniquely determine gamma parameters a and b that have
percentile values at the right locations
We call this method of specifying gamma prior parameters the

50/95 method (or the 50/90 method, if we use MTBF10 , etc.). A
simple way to calculate a and b for this method, using EXCEL, is
described below.
iii) A third way of choosing prior parameters starts the same way as the
second method. Consensus is reached on an reasonable MTBF, MTBF50.
Next, however, the group decides they want a somewhat weak prior that
will change rapidly, based on new test information. If the prior parameter
"a" is set to 1, the gamma has a standard deviation equal to its mean,
which makes it spread out, or "weak". To insure the 50th percentile is set
at 50 = 1/ MTBF50 , we have to choose b = ln 2 × MTBF50, which is
approximately .6931 × MTBF50.

Note: As we will see when we plan Bayesian tests, this weak prior is
actually a very friendly prior in terms of saving test time
Many variations are possible, based on the above three methods. For example,
you might have prior data from sources that you don't completely trust. Or you
might question whether the data really applies to the system under
investigation. You might decide to "weight" the prior data by .5, to "weaken" it.
This can be implemented by setting a = .5 x the number of fails in the prior data
and b = .5 times the number of test hours. That spreads out the prior distribution
more, and lets it react quicker to new test data.
Consequences
After a new No matter how you arrive at values for the gamma prior parameters a and b, the
test is run, method for incorporating new test information is the same. The new
the information is combined with the prior model to produce an updated or
posterior
posterior distribution modelfor .
gamma
parameters Under assumptions 1 and 2, when a new test is run with T system operating
are easily
obtained hours and r failures, the posterior distribution for is still a gamma, with new
from the parameters:
prior a' = a + r, b' = b + T
parameters
by adding In other words, add to a the number of new failures and add to b the number of
the new new test hours to get the new parameters for the posterior distribution.
number of
Use of the posterior distribution to estimate the system MTBF (with confidence,
fails to "a"
or prediction, intervals) is described in the section on estimating reliability
and the new
test time to using the Bayesian gamma model.
"b" Using EXCEL To Obtain Gamma Parameters
EXCEL can We will describe how to obtain a and b for the 50/95 method and indicate the
easily solve minor changes needed when any 2 other MTBF percentiles are used. The step
for gamma by step procedure is
prior 1. Calculate the ratio RT = MTBF50/MTBF05.
parameters
when using 2. Open an EXCEL spreadsheet and put any starting value guess for a in A1
the "50/95" - say 2.
consensus
method 3. Move to B1 and type the following expression:
= GAMMAINV(.95,A1,1)/GAMMAINV(.5,A1,1)
4. Press enter and a number will appear in B1. We are going to use the

"Goal Seek" tool EXCEL has to vary A1 until the number in B1 equals
RT.
5. Click on "Tools" (on the top menu bar) and then on "Goal Seek". A box
will open. Click on "Set cell" and highlight cell B1. $B$1 will appear in
the "Set Cell" window. Click on "To value" and type in the numerical
value for RT. Click on "By changing cell" and highlight A1 ($A$1 will
appear in "By changing cell"). Now click "OK" and watch the value of
the "a" parameter appear in A1.
6. Go to C1 and type
= .5*MTBF50*GAMMAINV(.5, A1, 2)
and the value of b will appear in C1 when you hit enter.

Example
An EXCEL A group of engineers, discussing the reliability of a new piece of equipment,

example decide to use the 50/95 method to convert their knowledge into a Bayesian
using the gamma prior. Consensus is reached on a likely MTBF50 value of 600 hours and
"50/95" a low MTBF05 value of 250. RT is 600/250 = 2.4. The figure below shows the
consensus EXCEL 5.0 spreadsheet just prior to clicking "OK" in the "Goal Seek" box.
method
After clicking "OK", the value in A1 changes from 2 to 2.862978. This new

value is the prior a parameter. (Note: if the group felt 250 was a MTBF10 value,
instead of a MTBF05 value, then the only change needed would be to replace
0.95 in the B1 equation by 0.90. This would be the "50/90" method.)
The figure below shows what to enter in C1 to obtain the prior "b" parameter
value of 1522.46.
The gamma prior with parameters a = 2.863 and b = 1522.46 will have
(approximately) a probability of 50% of λ being below 1/600 = .001667 and a
probability of 95% of being below 1/250 = .004. This can be checked by
typing
=GAMMDIST(.001667,2.863,(1/1522.46), TRUE)
and
=GAMMDIST(.004,2.863,(1/1522.46), TRUE)
as described when gamma EXCEl functions were introduced in Section 1.
This example will be continued in Section 3, where the Bayesian test time
needed to confirm a 500 hour MTBF at 80% confidence, will be derived.

8.3. Reliability Data Collection

In order to assess or improve reliability, it is usually necessary to have
failure data. Failure data can be obtained from field studies of system
performance or from planned reliability tests, sometimes called Life
Tests. This section focuses on how to plan reliability tests. The aim is to
answer questions such as: how long should you test, what sample size
do you need and what test conditions or stresses need to be run?
Detailed The section detailed outline follows.

contents of
Section 8.3 3. Reliability Data Collection
1. How do you plan a reliability assessment test?
1. Exponential life distribution (or HPP model) tests
2. Lognormal or Weibull tests
3. Reliability growth tests (Duane model)
4. Accelerated life tests
5. Bayesian gamma prior model tests

8.3.1. How do you plan a reliability assessment test?

8.3.1. How do you plan a reliability

assessment test?
Planning a Planning a reliability test means:
reliability ● How long should you test?
test ends
● How many units have to be put on test?
with a
detailed ❍ For repairable systems, this is often limited to 1.
description ● If acceleration modeling is part of the experimental plan

of the What combination of stresses and how many experimental
❍
mechanics cells?
of the test
❍ How many units go in each cell?
and starts
with stating The answers to these questions depend on:
your ● What models you are assuming?
assumptions ● What decisions or conclusions do you want to make after running
and what the test and analyzing the data?
you want to
● What risks you are willing to take of making wrong decisions or
discover or
prove conclusions?
It is not always possible, or practical, to completely answer all these
questions for every model we might want to use. This section looks at
answers, or guidelines, for the following models:
● exponential or HPP Model
● Weibull or lognormal model

● Duane or NHPP Power Law model
● acceleration models
● Bayesian gamma prior model

8.3.1.1. Exponential life distribution (or HPP model) tests

8.3.1.1. Exponential life distribution (or HPP

model) tests
Using an Exponential tests are common in industry for verifying that tools,
exponential systems or equipment are meeting their reliability requirements for
(or HPP) Mean Time Between Failure (MTBF). The assumption is that the system
model to test has a constant failure (or repair) rate, which is the reciprocal of the
whether a MTBF. The waiting time between failures has the exponential
system distribution model.
meets its
MTBF A typical test situation might be: a new complex piece of equipment or
requirement tool is installed in a factory and monitored closely for a period of several
is common weeks to several months. If it has no more than a pre specified number
in industry of failures during that period, the equipment "passes" its reliability
acceptance test.
This kind of reliability test is often called a Qualification Test or a
Product Reliability Acceptance Test (PRAT). Contractual penalties
may be invoked if the equipment fails the test. Everything is pegged to
meeting a customer MTBF requirement at a specified confidence level.
How Long Must You Test A Piece of Equipment or a System In
order to Assure a Specified MTBF at a Given Confidence?
You start with a given MTBF objective, say M, and a confidence level,
say 100 × (1- ). You need one more piece of information to determine
the test length: how many fails do you want to allow and still "pass" the
equipment? The more fails allowed, the longer the test required.
However, a longer test allowing more failures has the desirable feature
of making it less likely a good piece of equipment will be rejected
because of random "bad luck" doing the test period.
The recommended procedure is to iterate on r = the number of allowable
fails, until a larger r would require an unacceptable test length. For any
choice of r, the corresponding test length is quickly calculated by
multiplying M (the objective) by the factor in the table below

corresponding to the r-th row and the desired confidence level column.
For example, to confirm a 200 hour MTBF objective at 90% confidence,
allowing up to 4 failures on the test, the test length must be 200 × 7.99 =
1598 hours. If this is unacceptably long, try allowing only 3 fails for a
test length of 200 × 6.68 = 1336 hours. The shortest test would allow no
fails and last 200 × 2.3 = 460 hours. All these tests guarantee a 200 hour
MTBF at 90% confidence, when the equipment passes. However, the
shorter test are much less "fair" to the supplier, in that they have a large
chance of failing a marginally acceptable piece of equipment.
Use the Test Test Length Guide Table

length Table
to determine NUMBER
how long to OF
FACTOR FOR GIVEN CONFIDENCE LEVELS
FAILURES
test ALLOWED
r 50% 60% 75% 80% 90% 95%

0 .693 .916 1.39 1.61 2.30 3.00
1 1.68 2.02 2.69 2.99 3.89 4.74
2 2.67 3.11 3.92 4.28 5.32 6.30
3 3.67 4.18 5.11 5.52 6.68 7.75
4 4.67 5.24 6.27 6.72 7.99 9.15
5 5.67 6.29 7.42 7.90 9.28 10.51
6 6.67 7.35 8.56 9.07 10.53 11.84
7 7.67 8.38 9.68 10.23 11.77 13.15
8 8.67 9.43 10.80 11.38 13.00 14.43
9 9.67 10.48 11.91 12.52 14.21 15.70
10 10.67 11.52 13.02 13.65 15.40 16.96
15 15.67 16.69 18.48 19.23 21.29 23.10
20 20.68 21.84 23.88 24.73 29.06 30.89
The formula to calculate the factors in the table is:
and a Dataplot expression to calculate test length factors is

Dataplot LET FAC = .5*CHSPPF([1- ],[2*(r+1)])

expression
for The equivalent EXCEL expression for FAC is
obtaining
same factors = .5* CHIINV( ,(2*(r+1))).
as in Table Example: A new factory tool must meet a 400 hour MTBF requirement
at 80% confidence. You have up to two months of 3 shift operation to
decide whether the tool is acceptable. What is a good test plan?
Two months of around the clock operation, with some time off for
maintenance and repairs, amounts to a maximum of about 1300 hours.
The 80% confidence factor for r = 1 is 2.99, so a test of 400 × 2.99 =
about 1200 hours (with up to 1 fail allowed) is the best that can be
done.
Shorten NOTE: Exponential test times can be shortened significantly if several

required test similar tools or systems can be put on test at the same time. Test time
times by means the same as "tool hours" and 1 tool operating 1000 hours is
testing more equivalent (as far as the exponential model is concerned) to 2 tools
than 1 operating for 500 hours each or 10 tool operating for 100 hours each.
system Just count all the fails from all the tools and the sum of the test hours
from all the tools.

8.3.1.2. Lognormal or Weibull tests


Planning Planning a reliability test is not simple and straightforward when the
reliability assumed model is lognormal or Weibull. Since these models have two
tests for parameters, no estimates are possible without at least two test failures,
distributions and good estimates require considerably more than that. Because of
other than censoring, without a good guess ahead of time at what the unknown
the parameters are, any test plan may fail.
exponential
is difficult However, it is often possible to make a good guess ahead of time about
and involves at least one of the unknown parameters - typically the "shape"
a lot of parameter ( for the lognormal or for the Weibull). With one
guesswork parameter assumed known, test plans can be derived that assure the
reliability or failure rate of the product tested is acceptable.
Lognormal Case (shape parameter known): The lognormal model is
used for many microelectronic wear out failure mechanisms, such as
electromigration. As a production monitor, samples of microelectronic
chips taken randomly from production lots might be tested at levels of
voltage and temperature that are high enough to significantly accelerate
the occurrence of electromigration failures. Acceleration factors are
known from previous testing and range from several hundred to several
thousand.

Lognormal The goal is to come up with a test plan (put n units on stress test for T
test plans, hours and accept the lot if no more than r failures occur). The following
assuming assumptions are made:
sigma and ● The life distribution model is lognormal
the
acceleration. ● Sigma = is known from past testing and does not vary
Factor is appreciably from lot to lot
known ● Lot reliability varies because T50's (the lognormal median or 50th
percentile) differ from lot to lot
● The acceleration factor from high stress to use stress is a known
quantity "A"
● A stress time of T hours is practical as a line monitor
● A nominal use T50 of Tu (combined with ) produces an

acceptable use CDF (or use reliability function). This is
equivalent to specifying an acceptable use CDF at, say, 100,000
hours to be a given value p0 and calculating Tu via:
where is the inverse of the standard normal distribution

● An unacceptable use CDF of p1 leads to a "bad" use T50 of Tb ,
using the same equation as above with po replaced by p1
The acceleration factor A is next used to calculate a "good" or
acceptable proportion of failures pa at stress and a "bad" or unacceptable
proportion of fails pb:
where is the standard normal CDF. This reduces the reliability

problem to a well known Lot Acceptance Sampling Plan (LASP)
problem, which was covered in Chapter 6.
If the sample size required to distinguish between pa and pb turns out to

be too large, it may be necessary to increase T or test at a higher stress.
The important point is that the above assumptions and equations give a
methodology for planning ongoing reliability tests under a lognormal
model assumption.

Weibull test Weibull Case (shape parameter known): The assumptions and
plans, calculations are similar to those made for the lognormal:
assuming ● The life distribution model is Weibull
gamma and
the ● Gamma = is known from past testing and does not vary
acceleration. appreciably from lot to lot
factor is
known ● Lot reliability varies because 's (the Weibull characteristic life
or 62.3 percentile) differ from lot to lot
● The acceleration factor from high stress to use stress is a known
quantity "A"
● A stress time of T hours is practical as a line monitor
● A nominal use of u (combined with ) produces an
acceptable use CDF (or use reliability function). This is
equivalent to specifying an acceptable use CDF at, say, 100,000
hours to be a given value p0and calculating u
● An unacceptable use CDF of p1 leads to a "bad" use of of ,

using the same equation as above with po replaced by p1
The acceleration factor A is used next to calculate a "good" or
acceptable proportion of failures pa at stress and a "bad" or
unacceptable proportion of fails pb:
This reduces the reliability problem to a Lot Acceptance Sampling Plan

(LASP) problem, which was covered in Chapter 6.
If the sample size required to distinguish between pa and pb turns out to

be too large, it may be necessary to increase T or test at a higher stress.
The important point is that the above assumptions and equations give a
methodology for planning ongoing reliability tests under a Weibull
model assumption.
Planning Tests to Estimate Both Weibull or Lognormal Parameters

Rules of All that can be said here are some general rules of thumb:
thumb for 1. If you can observe at least 10 exact times of failure, estimates are
general usually reasonable - below 10 fails the critical shape parameter
lognormal may be hard to estimate accurately. Below 5 failures, estimates
or Weibull are often very inaccurate.
life test
2. With readout data, even with more than 10 total failures, you
planning
need failures in three or more readout intervals for accurate
estimates.
3. When guessing how many units to put on test for how long, try
out various reasonable combinations of distribution parameters to
see if the corresponding calculated proportion of failure expected
during the test, multiplied by the sample size, gives a reasonable
number of failures.
4. As an alternative to the last rule, simulate test data from
reasonable combinations of distribution parameters and see if
your estimates from the simulated data are close to the parameters
used in the simulation. If a test plan doesn't work well with
simulated data, it is not likely to work well with real data.

8.3.1.3. Reliability growth (Duane model)


Guides for A reliability improvement test usually takes a large resource
planning commitment, so it is important to have a way of estimating how long a
how long to test will be required. The following procedure gives a starting point for
run a determining a test time:
reliability 1. Guess at a starting value for , the growth slope. Some
growth test
guidelines were previously discussed. Pick something close to .3
for a conservative estimate (perhaps a new cross functional team
will be working on the improvement test or the system to be
improved has many new parts with possibly unknown failure
mechanisms) or up around .5 for a more optimistic estimate.
2. Use current data and engineering estimates to arrive at a
consensus of what the starting MTBF for the system is. Call this
M1.
3. Let MT be the target MTBF (the customer requirement). Then the
improvement needed on the test is given by
IM = MT/M1
4. A first pass estimate of the test time needed is
This estimate comes from placing the starting MTBF of M1 to be the

MTBF after 1 hour on test and using the fact that the improvement from
1 hour to T hours is just

Make sure The reason the above is just a first pass estimate is it will give
test time unrealistic (too short) test times when a high is assumed. A very
makes
short reliability improvement test makes little sense because a minimal
engineering
number of failures must be observed before the improvement team can
sense
detrmine design and parts changes that will "grow" reliability. And it
takes time to implement these changes and observe an improved repair
rate.
Iterative Simulation methods can also be used to see if a planned test is likely to
simulation is generate data that will demonstrate an assumed growth rate.
an aid for
test
planning

8.3.1.4. Accelerated life test


Accelerated Accelerated life tests are component life tests where components are
testing is operated at high stresses and failure data is observed. Often these tests
needed when are done for two purposes:
testing even 1. To study how failure is accelerated by stress and fit an
large sample acceleration model to data from multiple stress cells
sizes at use
2. To obtain enough failure data at high stress to accurately project
stress would
(extrapolate) what the CDF at use will be.
yield few or
no failures If we already know the acceleration model (or the acceleration factor
in a from high stress, where we can observe a reasonable number of failures,
reasonable and use stress, where we want to confirm that customer requirements
time will be met), then the methods described in the last section can be used.
We assume, therefore, that the acceleration model is not known in
advance.
Test Test planning and operation for a (multiple) stress cell life test
planning experiment consists of the following:
means ● Pick several combinations of the relevant stresses (the stresses
picking that accelerate the failure mechanism under investigation). Each
stress levels combination is a "stress cell". Note that you are planning for only
and sample one mechanism of failure at a time. Failures on test due to any
sizes and other mechanism will be considered censored run times.
test times to
produce ● Make sure stress levels used are not too high - to the point where
enough data new failure mechanisms that would never occur at use stress are
to fit models introduced. Picking a maximum allowable stress level requires
and make experience and/or good engineering judgment.
projections ● Put random samples of components in each stress cell and run the
components in each cell for fixed (but possibly different) lengths
of time.
● Gather the failure data from each cell and use it to fit an
acceleration model and a life distribution model and use these
models to project reliability at use stress conditions.

Test planning would be similar to topics already covered in the chapters

that discussed modeling and experimental design except for one
important point. When you test components in a stress cell for a fixed
length test, it is typical that some (or possibly many) of the components
end the test without failing. This is the censoring problem, and it greatly
complicates experimental design to the point where it becomes almost
as much of an art (based on engineering judgment) as a statistical
science.
A example will help illustrate the design issues. Assume a metal

migration failure mode is believed to follow the 2-stress temperature
voltage model given by
Normal use conditions are 4 volts and 25 degrees celsius, and the high
stress levels under consideration are 6, 8,12 volts and 85o, 105o and
125o. It probably would be a waste of resources to test at (6v, 85o), or
even possibly (8v, 85o) or (6v,105o), since these cells are not likely to
have enough stress acceleration to yield a reasonable number of failures
within typical test times.
If you write all the 9 possible stress cell combinations in a 3x3 matrix
with voltage increasing by rows and temperature increasing by columns,
the result would look like the matrix below:
Matrix Leading to "Backward L Design"
6v, 85o 6v, 105o 6v, 125o
8v, 85o 8v,105o 8v,125o
12v,85o 12v,105o 12v,125o

"Backwards The combinations in bold are the most likely design choices covering
L" designs the full range of both stresses, but still hopefully having enough
are common acceleration to produce failures. This is the so-called "backwards L"
in design commonly used for acceleration modeling experiments.
accelerated
life testing. Note: It is good design practice to put more of your test units in the
Put more lower stress cells, to make up for the fact that these cells will have a
experimental smaller proportion of units failing.
units in
lower stress
cells.
Sometimes Design by Simulation:

simulation is
the best way A lengthy, but more accurate way to choose a test matrix is the
to learn following:
whether a ● Pick an acceleration model and a life distribution model (as
test plan has usual)
a chance of ● Guess at the shape parameter of the life distribution model based
working on literature studies or earlier experiments. The shape parameter
should remain the same for all stress cells. Choose a scale
parameter at use so that the use stress CDF exactly meets
requirements (i.e., for the lognormal, pick a use T50 that gives the
desired use reliability - for a Weibull model choice, do the same
for the characteristic life parameter)
● Guess at the acceleration model parameters ( H and
, for the 2 stress model shown above). Again, use whatever is
in the literature for similar failure mechanisms or data from
earlier experiments).
● Calculate acceleration factors from any proposed test cells to use
stress and divide the use scale parameter by these acceleration
factors to obtain "trial" cell scale parameters.
● Simulate cell data for each proposed stress cell using the derived
cell scale parameters and the guessed shape parameter.
● Check that every proposed cell has sufficient fails to give good
estimates.
● Adjust the choice of stress cells and the sample size allocations
until you are satisfied that, if everything goes as expected, the
experiment will yield enough data to accurately estimate model
parameters.

After you Optimal Designs:

make
advance Recent work on designing accelerated life tests has shown it is possible,
guesses, it is for a given choice of models and assumed values of the unknown
sometimes parameters, to come up with an optimal design (one which will have the
possible to best chance of giving good sample estimates of the model parameters).
come up These optimal models typically pick stress levels as far apart as possible
with an and heavily weight the allocation of sample units to the lower stress
optimal cells. However, unless the experimenter can find software that
experimental incorporates these optimal methods for his or her particular choice of
design - but models, the methods described above are the most practical way of
software for designing acceleration experiments.
this is rare

8.3.1.5. Bayesian gamma prior model


How to Review Bayesian basics and assumptions, if needed. We start at the point
plan a where gamma prior parameters a and b have already been determined.
Bayesian Assume we have a given MTBF objective, say M, and a desired confidence
test to level, say 100×(1- ). We want to confirm the system will have at least an
confirm a
MTBF of M at the 100×(1- ) confidence level. As in the section on
system
classical (HPP) test plans, we pick a number of failures, r, we can allow on
meets its
MTBF the test. We need a test time T such that we can see up to r failures, and
objective still "pass" the test. If the test time is too long (or too short), we can iterate
with a different choice of r.
When the test ends, the posterior gamma distribution will have (worst case
- assuming exactly r failures) new parameters of
a' = a + r, b' = b + T
and passing the test means that the failure rate the upper 100×(1- )
percentile for the posterior gamma, has to equal the target failure rate 1/M.
But this percentile is, by definition, G-1(1- ;a',b'), where G-1 is the
inverse of the gamma distribution with parameters a', b'. We can find the
value of T that satisfies G-1(1- ;a',b'), = 1/M} by trial and error, or by
using "Goal Seek" in EXCEL. However, based on the properties of the
gamma distribution, it turns out that we can calculate T directly, by using
T = .5M × G-1 (1- ; 2a',.5) - b

Excel will Solving For T = Bayesian Test Time Using EXCEL or Dataplot
easily do
the The EXCEL expression for the required Bayesian test time to confirm a
required goal of M at 100×(1-a)% confidence, allowing r failures and assuming
calculations gamma prior parameters of a and b is
= .5*M*GAMMAINV((1- ),((a+r)),2) - b
and the equivalent Dataplot expression is
LET BAYESTIME = M*GAMPPF((1- ),(a+r)) - b.
Special Case: The Prior Has a = 1 (The "Weak" Prior)
When the There is a very simple way to calculate the required Bayesian test time,
prior is a when the prior is a weak prior with a = 1. Just use the Test Length Guide
weak prior Table to calculate the classical test time. Call this Tc. The Bayesian test
with a = 1, time T is just Tc minus the prior parameter b (i.e. T = Tc - b). If the b
the
parameter was set equal to (ln 2) × MTBF50 (where MTBF50 is the
Bayesian
test is consensus choice for an "even money" MTBF), then
always T = Tc - (ln 2) x MTBF50
shorter This shows that when a weak prior is used, the Bayesian test time is always
than the less then the corresponding classical test time. That is why this prior is also
classical known as a friendly prior.
test
Note: In general, Bayesian test times can be shorter, or longer, than the
corresponding classical test times, depending on the choice of prior
parameters. However, the Bayesian time will always be shorter when the
prior parameter a is less than, or equal to, 1.
Example: Calculating a Bayesian Test Time
EXCEL A new piece of equipment has to meet a MTBF requirement of 500 hours
example at 80% confidence. A group of engineers decide to use their collective
experience to come up with a Bayesian gamma prior using the 50/95
method described in Section 2. They think 600 hours is a likely MTBF
value and they are very confident that the MTBF will exceed 250.
Following the example in Section 2, they determine that the gamma prior
parameters are a = 2.863 and b = 1522.46.
Now they want to determine an appropriate test time, so that they can
confirm a MTBF of 500 with at least 80% confidence, provided they have
no more than 2 failures.
Using an EXCEL spreadsheet, type the expression
= .5*500*GAMMAINV(.8,((a+r)),2) - 1522.46

and the required test time of 1756 hours will appear (as shown below).
Using Dataplot, the same result would be obtained from

LET BAYESTIME = 500*GAMPPF(.8,4.863) - 1522.46
To compare this result to the classical test time required, use the Test
Length Guide Table. The table factor is 4.28, so the test time needed is 500
x 4.28 = 2140 hours for a non Bayesian test. The Bayesian test saves about
384 hours, or an 18% savings. If the test is run for 1756 hours, with no
more than 2 failures, then (at least) an MTBF of 500 has been confirmed at
80% confidence.
If, instead, the engineers had decided to use a weak prior with an MTBF50
of 600, the required test time would have been
2140 - 600 x ln 2 = 1724 hours

8.4. Reliability Data Analysis

After you have obtained component of system reliability data, how do
you fit life distribution models, reliability growth models, or
acceleration models? How do you estimate failure rates or MTBF's and
project component or system reliability at use conditions? This section
answers these kinds of questions.
Detailed The detailed outline for section 4 follows.

outline for
Section 4 4. Reliability Data Analysis
1. How do you estimate life distribution parameters from censored
data?
1. Graphical estimation
2. Maximum Likelihood Estimation (MLE)
3. A Weibull MLE example
2. How do you fit an acceleration model?
1. Graphical estimation
2. Maximum likelihood
3. Fitting models using degradation data instead of failures
3. How do you project reliability at use conditions?
4. How do you compare reliability between two or more
populations?
5. How do you fit system repair rate models?
1. Constant repair rate (HPP/Exponential) model
2. Power law (Duane) model
3. Exponential law model
6. How do you estimate reliability using the Bayesian gamma prior
model?


8.4.1. How do you estimate life distribution parameters from censored Data?

8.4.1. How do you estimate life distribution

parameters from censored Data?
Graphical Two widely used general methods will be described in this section:
estimation ● Graphical estimation
methods
(aided by ● Maximum Likelihood Estimation (MLE)
computer Recommendation On Which Method to Use
line fits) are
easy and Maximum likelihood estimation (except when the failure data is very
quick. sparse - i.e. only a few failures) is a more accurate and flexible method.
However, with censored data, the method of maximum likelihood
Maximum estimation requires special computer programs for distributions other
likelihood than the exponential. It may be difficult or costly to obtain MLE
methods are programs that meet your needs.
usually
more If important business decisions are based on reliability projections made
accurate - from life test data and acceleration modeling, then it pays to search for
but need and obtain state of the art MLE reliability software. Otherwise, for
special monitoring and tracking reliability, estimation methods based on
software computer augmented graphical procedures will generally suffice.

8.4.1.1. Graphical estimation


Every line Once you have calculated plotting positions from your failure data, and
on put the points on the appropriate graph paper for your chosen model,
probability parameter estimation follows easily. But along with the mechanics of
paper graphical estimation, be aware of both the advantagesand the
uniquely
disadvantages of graphical estimation methods.
identifies
distribution
parameters
Most Graphical Estimation Mechanics:

probability
papers have If you draw a line through the points, and the paper is a commercially
simple designed probability paper, there are usually simple rules to find the
procedures slope (or shape parameter) and the scale parameter. On lognormal paper
than go with time on the x-axis and cum percent on the y-axis, draw horizontal
from a line lines from the 34th and the 50th percentile across to the line, and drop
to the vertical lines to the time axis from these intersection points. The time
underlying corresponding to the 50th percentile is the T50 estimate. Divide T50 by
distribution the time corresponding to the 34th percentile (this is called T34). The
parameters natural logarithm of that ratio is the estimate of sigma, or the slope of
the line ( = ln ( T50 / T34)).
On commercial Weibull probability paper, there is often a horizontal

line through the 62.3 percentile point. That estimation line intersects the
line through the points at a time that is the characteristic life parameter
. In order to estimate the line slope (or the shape parameter ), some
papers have a special point on them called an estimation point. You drop
a line from the estimation point perpendicular to the fitted line and look
at where it passes through a special estimation scale. The value of is
read off the estimation scale where the line crosses it.
Other papers may have variations on the methods described above.

Using a To remove the subjectivity of drawing a line through the points, a least
computer squares (regression) fit can be done using the equations described in the
generated section on how special papers work. An example of this for the Weibull,
line fitting using the Dataplot FIT program, was also shown in that section. A SAS
routine JMP™ example of a Weibull plot for the same data is shown later in
removes this section.
subjectivity
and can Finally, if you have exact times and complete samples (no censoring),
lead directly Dataplot has built in Probability Plotting functions and a built in
to computer Weibull paper - examples were shown in the sections on the various life
parameter distribution models.
estimates
based on the
plotting
positions
Do Advantages of Graphical Methods of Estimation:

probability ● Graphical methods are quick and easy to use and make visual
plots even if sense
you use
● Calculations can be done with little or no special software needed.
some other
method for ● Visual test of model (i.e. how well the points line up) is an
the final additional benefit

estimates Disadvantages of Graphical Methods of Estimation
Perhaps the The statistical properties of graphical estimates (i.e. how accurate are
worst they on the average) are not good
drawback of ● they are biased
graphical
● even with large samples, they do not become minimum variance
estimation is
(i.e. most accurate) estimates
you cannot
get ● graphical methods do not give confidence intervals for the
legitimate parameters (intervals generated by a regression program for this

confidence kind of data are incorrect)
intervals for ● Formal statistical tests about model fit or parameter values can
the not be done with graphical methods
estimates As we will see in the next section, Maximum Likelihood Estimates
overcome all these disadvantages - at least for reliability data sets with a
reasonably large number of failures - at a cost of losing all the
advantages listed above for graphical estimation.


8.4.1.2. Maximum likelihood estimation


There is Maximum likelihood estimation begins with writing a mathematical
nothing expression known as the Likelihood Function of the sample data.
visual about Loosely speaking, the likelihood of a set of data is the probability of
the obtaining that particular set of data, given the chosen probability
maximum distribution model. This expression contains the unknown model
likelihood parameters. Those values of these parameters that maximize the sample
method - but likelihood are known as the Maximum Likelihood Estimates or
it is a MLE's.
powerful
method and, Maximum likelihood estimation is a totally analytic maximization
at least for procedure. It applies to every form of censored or multicensored data,
large and it is even possible to use the technique across several stress cells and
samples, estimate acceleration model parameters at the same time as life
very distribution parameters. Moreover, MLE's and Likelihood Functions
accurate generally have very desirable large sample properties:
● they become unbiased minimum variance estimates as the sample
size increases
● they have approximate normal distributions and approximate
sample variances that can be calculated and used to generate
confidence bounds
● likelihood functions can be used to test hypotheses about models
and parameters

With small There are only two drawbacks to MLE's, but they are important ones:
samples, ● With small numbers of failures (less than 5, and sometimes less
MLE's may than 10 is small), MLE's can be heavily biased and the large
not be very sample optimality properties do not apply
accurate
● Calculating MLE's often requires solving complex non-linear
and may
equations requiring specialized software. This is less of a problem
even
as time goes by, as more statistical packages are upgrading to
generate a
have MLE analysis capability every year.
line that lies
above or
below the
data points
Likelihood Likelihood Function Examples for Reliability Data:

equation for
censored Let f(t) be the PDF and F(t) the CDF for the chosen life distribution
data model. Note that these are functions of t and the unknown parameters of
the model. The likelihood function for Type I Censored data is:
where C is a constant that plays no role when solving for the MLE's.
Note that with no censoring, the likelihood reduces to just the product of
the densities, each evaluated at a failure time. For Type II Censored
Data, just replace T above by the random end of test time tr.
The likelihood function for readout data is:
with F(T0) defined to be 1.
In general, any multicensored data set likelihood will be a constant

times a product of terms, one for each unit in the sample, that look like
either f(ti), [F(Ti)-F(Ti-1)], or [1-F(ti)], depending on whether the unit
was an exact time failure at time ti, failed between two readouts Ti-1 and
Ti, or survived to time ti and was not observed any longer.
The general mathematical technique for solving for MLE's involves

setting partial derivatives of ln L (the derivatives are taken with respect

to the unknown parameters) equal to zero and solving the resulting

(usually non-linear) equations. The exponential model, however, can
easily be solved.
MLE for the MLE's for the Exponential Model (Type I Censoring):
exponential
model
parameter
turns out
to be just
(total # of
failures)
divided by
(total unit
test time)
Note: The MLE of the failure rate (or repair rate) in the exponential case
turns out to be the total number of failures observed divided by the total
unit test time. For the MLE of the MTBF, take the reciprocal of this or
use the total unit test hours divided by the total observed failures.
There are examples of Weibull and lognormal MLE analysis, using SAS
JMP™ software, later in this section.

8.4.1.3. A Weibull maximum likelihood estimation example

8.4.1.3. A Weibull maximum likelihood estimation

example
Reliability SAS JMP™ Example
analysis
with a SAS JMP software has excellent survival analysis (i.e. reliability analysis) capabilities,
popular incorporating both graphical plotting and maximum likelihood estimation and covering
statistical the exponential, Weibull, lognormal and extreme value distribution models.
software Use of JMP (release 3) for plotting Weibull censored data and estimating parameters will
package be illustrated using data from a previous example.
Steps in a Weibull Data Example

Weibull
analysis Failure times were 55, 187, 216, 240, 244, 335, 361, 373, 375, and 386 hours and 10
using JMP unfailed units were removed from test at 500 hours. The steps in creating a JMP
software worksheet and analyzing the data are as follows:
1. Set up three columns, one for the failure and censoring times ("Time"), another to
indicate whether the time is a failure or a censoring time ("Cens") and the third column
to show how many units failed or were censored at that time ("Freq"). Fill in the 11 times
above, using "0" in Cens to indicate a failure and "1" in Cens to indicate a censoring
time. The spreadsheet will look as follows:
You can obtain a copy of this JMP worksheet by clicking here mleex.jmp (if your

browser is configured to bring up JMP automatically, you can try out the example as you
read about it).
2. Click on Analyze, choose "Survival" and then choose "Kaplan - Meier Method". Note:
Some software packages (and other releases of JMP) might use the name "Product Limit
Method" or "Product Limit Survival Estimates", instead of the equivalent name
"Kaplan-Meier".
3. In the Box that appears, select the columns from mleex that correspond to "Time",
"Censor" and "Freq", put them in the corresponding slots on the right (see below) and
click "OK".
4. Click "OK" and the analysis results appear. You may have to use the "check mark" tab
on the lower left to select Weibull Plot (other choices are Lognormal and Exponential).

You may also have to open the tab next to the words "Weibull Plot" and select "Weibull
Estimates". The results are shown below.
Note: JMP uses the parameter for the Weibull characteristic life (as does Dataplot),
and the parameter for the shape (Dataplot uses ). The Extreme Value distribution
parameters estimates are for the distribution of "ln time to fail" and have the relationship
5. There is an alternate way to get some of the same results that can also be used to fit
models when there are additional "effects" such as temperature differences or vintage or
plant of manufacture differences. Instead of clicking "Kaplan - Meier Method" in step 2,
chose "Parametric Model" after selecting "Survival" from the "Analysis" choices. The
screen below appears. Repeat step 3 and make sure "Weibull" appears as the "Get
Model" choice. In this example, there are no other effects to "Add" ( the acceleration
model example later on will illustrate how to add a temperature effect). Click "Run
Model" to get the results below. This time, you need to use the check symbol tab to

obtain confidence limits. Only the Extreme Value distribution parameter estimates are
displayed.

Limitations Notes:
and a
warning 1. The built in reliability analysis routine that are currently a part of JMP only handle
about the exact time of failure data with possible right censoring. However, use of templates
Likelihood (provided later in the Handbook) for either Weibull or lognormal data, extends JMP
calculation analysis capabilities to handle readout (interval) data and any type of censoring or
in JMP truncation. This will be described in the acceleration model example later on.
2. The "Model Fit" screen for the Weibull model gives a value for -Loglikelihood for the
Weibull fit. This should be the negative of the maximized likelihood function. However,
JMP leaves out a term consisting of the sum of all the ln time of failures in the data set.
This does not affect the calculation of MLE's or confidence bounds but can be confusing
when comparing results between different software packages. In the example above, the
sum of the ln times is ln 55 + ln 187 + . . . + ln 386 = 55.099 and the correct maximum
log likelihood is - (20.023 + 55.099) = - 75.122.
3. The omission of the sum of the ln times of failure in the likelihood also occurs when
fitting lognormal and exponential models.
4. Different releases of JMP may, of course, operate somewhat differently. The analysis
shown here used release 3.2.2.
Conclusions
MLE analysis is an accurate and easy way to estimate life distribution parameters,
provided a good software analysis package is available. The package should also
calculate confidence bounds and loglikelihood values. JMP has this capability, as well as
several other commercial statistical analysis packages.

8.4.2. How do you fit an acceleration model?

8.4.2. How do you fit an acceleration

model?
Acceleration As with estimating life distribution model parameters, there are two
models can general approaches for estimating acceleration model parameters:
be fit by ● Graphical estimation (or computer procedures based on a
either graphical approach)
graphical
● Maximum Likelihood Estimation (an analytic approach based on
procedures
or maximum writing the likelihood of all the data across all the cells,
likelihood incorporating the acceleration model).
methods The same comments and recommendations concerning these methods
still apply. Note that it is even harder, however, to find useful software
programs that will do maximum likelihood estimation across stress cells
and fit and test acceleration models.
Sometimes it Another promising method of fitting acceleration models is sometimes

is possible to possible when studying failure mechanisms based stress induced
fit a model gradual degradation processes. This approach fits models based on
using degradation data, and has the advantage of not actually needing failures.
degradation This overcomes censoring limitations by providing measurement data at
data consecutive time intervals for every unit in every stress cell.



This section will discuss the following:
1. How to fit an Arrhenius model with graphical estimation
2. Graphical estimation: an Arrhenius model example
3. Fitting more complicated models
Estimate How to fit an Arrhenius Model with Graphical Estimation

acceleration
model Graphical methods work best (and are easiest to describe) for a simple one stress model
parameters like the widely used Arrhenius model
by
estimating
cell T50's
(or 's)
and then where T is temperature measured in degrees Kelvin (273.16 + degrees Celsius) and k is
using Boltzmann's constant (8.617 x 10-5 in eV/°K).
regression
to fit the When applying an acceleration model to a distribution of failure times, we interpret the
model deterministic model equation to apply at any distribution percentile we want. This is
across the equivalent to setting the life distribution scale parameter equal to the model equation
cells (T50 for the lognormal, for the Weibull and the MTBF or 1/ for the exponential).
For the lognormal, for example, we have
So, if we run several stress cells and compute T50's for each cell, a plot of the natural log
of these T50's vs the corresponding 1/kT values should be roughly linear with a slope of
H and an intercept of ln A. In practice, a computer fit of a line through these points is
typically used to get the Arrhenius model estimates. There are even commercial
Arrhenius graph papers that have a temperature scale in 1/kT units and a T50 scale in log

units, but it is easy enough to make the transformations and then use linear or log linear
papers. Remember that T is in Kelvin in the above equations. For temperature in celsius,
use the following for 1/kT: 11605/(TCELSIUS + 273.16)
An example will illustrate the procedure.
Graphical Estimation: An Arrhenius Model Example:
Arrhenius Component life tests were run at 3 temperatures: 85°C, 105°C and 125°C. The lowest
model temperature cell was populated with 100 components; the 105° cell had 50 components
example and the highest stress cell had 25 components. All tests were run until either all the units
in the cell had failed or 1000 hours was reached. Acceleration was assumed to follow an
Arrhenius model and the life distribution model for the failure mode was believed to be
lognormal. The normal operating temperature for the components is 25°C, and it is
desired to project the use CDF at 100,000 hours.
Test results:
Cell 1 (85°C): 5 failures at: 401, 428, 695, 725 and 738 hours. 95 units were censored at
1000 hours running time.
Cell 2 (105°C): 35 failures at 171, 187, 189, 266, 275, 285, 301, 302, 305, 316, 317, 324,
349, 350, 386, 405, 480, 493, 530, 534, 536, 567, 589, 598, 599, 614, 620, 650, 668,
685, 718, 795, 854, 917, and 926 hours. 15 units were censored at 1000 hours running
time.
Cell3 (125°C): 24 failures at 24, 42, 92, 93, 141, 142, 143, 159, 181, 188, 194, 199, 207,
213, 243, 256, 259, 290, 294, 305, 392, 454, 502 and 696. 1 unit was censored at 1000
hours running time.
Failure analysis confirmed that all failures were due to the same failure mechanism (if
any failures due to another mechanism had occurred, they would have been considered
censored run times in the Arrhenius analysis).
Steps to Fitting the Distribution Model and the Arrhenius Model:
● Do graphical plots for each cell and estimate T50's and sigma's as previously
discussed.
● Put all the plots on the same sheet of graph paper and check whether the lines are
roughly parallel (a necessary consequence of true acceleration).
● If satisfied from the plots that both the lognormal model and the constant sigma
from cell to cell are consistent with the data, plot the cell ln T50's versus the
11605/(TCELSIUS + 273.16) cell values, check for linearity and fit a straight line
through the points. Since the points have different degrees of precision, because
different numbers of fails went into their calculation, it is recommended that the
number of fails in each cell be used as weights in a regression program, when
fitting a line through the points.
● Use the slope of the line as the H estimate and calculate the Arrhenius A
constant from the intercept using A = eintercept .
● Estimate the common sigma across all the cells by the weighted average of the
individual cell sigma estimates. Use the number of failures in a cell divided by the
total number of failures in all cells, as that cells weight. This will allow cells with
more failures to play a bigger role in the estimation process.

Dataplot Dataplot Analysis of Multicell Arrhenius Model Data:

solution for
Arrhenius After creating text files DAT1.TXT, DAT2.TXT and DAT3.TXT of the failure times for
model the 3 stress cells, enter Dataplot and execute the following sequence of commands
example (individual cell plots have been skipped):
READ DAT1.TXT CELL1
READ DAT2.TXT CELL2
READ DAT3.TXT CELL3
LET Y1 = LOG(CELL1)
LET Y2 = LOG(CELL2)
LET Y3 = LOG(CELL3)
LET POS1 = SEQUENCE 1 1 5
LET POS1 = (POS1 -.3)/100.4
LET POS2 = (POS2 -.3)/50.4
LET POS3 = (POS3 -.3)/25.4
LET X1 = NORPPF(POS1)
TITLE PROBABILITY PLOTS OF THREE TEMPERATURE CELLS
PLOT Y1 X1 AND
PLOT Y2 X2 AND
PLOT Y3 X3
This will produce the following probability plot of all three stress cells on the same
graph.

Note that the lines are reasonably straight (a check on the lognormal model) and the
slopes are approximately parallel (a check on the acceleration assumption).
The cell ln T50 and sigma estimates are obtained from the FIT function as follows:
FIT Y1 X1
FIT Y2 X2
FIT Y3 X3
Each FIT will yield a cell Ao = the ln T50 estimate and A1 = the cell sigma estimate.
These are summarized in the table below.
Summary of Least Squares Estimation of Cell Lognormal Parameters
Cell Number ln T50 Sigma
1 (T = 85) 8.168 .908
2 (T = 105) 6.415 .663
3 (T = 125) 5.319 .805
The three cells have 11605/(T + 273.16) values of 32.40, 30.69 and 29.15 in cell number
order. The Dataplot commands to generate the Arrhenius plot are:
LET YARRH = DATA 8.168 6.415 5.319
LET XARRH = DATA 32.4 30.69 29.15
TITLE = ARRHENIUS PLOT OF CELL T50'S
With only three cells, it is unlikely a straight line through the points will present obvious
visual lack of fit. However, in this case, the points appear to line up very well.
Finally, the model coefficients are computed from

LET SS = DATA 5 35 24
WEIGHT = SS
FIT YARRH XARRH
This will yield a ln A estimate of -18.312 (A = e-18.312 = .1115x10-7) and a H estimate
of .808. With this value of H, the acceleration between the lowest stress cell of 85°C
and the highest of 125°C is
which is almost 14× acceleration. Acceleration from 125 to the use condition of 25°C is
3708× . The use T50 is e-18.312 x e.808x11605x1/298.16= e13.137 = 507380.
A single sigma estimate for all stress conditions can be calculated as a weighted average
of the 3 sigma estimates obtained from the experimental cells. The weighted average is
(5/64) × .908 + (35/64) × .663 + (24/64) × .805 = .74.
Fitting More Complicated models
Models Two stress models, such as the temperature /voltage model given by
involving
several
stresses can
be fit using
multiple
regression need at least 4 or five carefully chosen stress cells to estimate all the parameters. The
Backwards L design previously described is an example of a design for this model. The
bottom row of the "backward L" could be used for a plot testing the Arrhenius
temperature dependence, similar to the above Arrhenius example. The right hand column
could be plotted using y = ln T50 and x = ln V, to check the voltage term in the model.
The overall model estimates should be obtained from fitting the multiple regression
model
The Dataplot command for fitting this model, after setting up the Y, X1 = X1, X2 = X2
data vectors, is simply
FIT Y X1 X2
and the output gives the estimates for b0, b1 and b2.
Three stress models, and even Eyring models with interaction terms, can be fit by a
direct extension of these methods. Graphical plots to test the model, however, are less
likely to be meaningful as the model gets more complex.


8.4.2.2. Maximum likelihood


This page This page contains the following sub-sections:
contains ● Description of MLE methods for acceleration models
descriptions
and step by ● Example comparing graphical estimates and MLE 's
step ● Using SAS JMPTM software to fit reliability models
examples of
● Steps for fitting the Arrhenius model using the "NonlinearFit" option and special
Maximum
Likelihood JMP templates
acceleration ● How to use JMP templates for lognormal or Weibull data (including acceleration
modeling model analysis)
● JMP template data entry for truncated or left censored Weibull or lognormal data
Description of MLE Methods for Acceleration Models
The The Likelihood equation for a multicell acceleration model starts by computing the
maximum Likelihood functions for each cell, as was described earlier. Each cell will have
likelihood unknown life distribution parameters that, in general, are different. For example, if a
method can lognormal model is used, each cell might have its own T50 and .
be used to
estimate Under an acceleration assumption, however, all the cells contain samples from
distribution populations that have the same value of (the slope does not change for different
and stress cells). Also, the T50's are related to one another by the acceleration model ; they
acceleration all can be written using the acceleration model equation with the proper cell stresses put
model in.
parameters
at the same To form the Likelihood equation under the acceleration model assumption, simply
time rewrite each cell Likelihood by replacing each cell T50 by its acceleration model
equation equivalent and replacing each cell sigma by the same one overall . Then,
multiply all these modified cell Likelihoods together to get the overall Likelihood
equation.
Once you have the overall Likelihood equation, the maximum likelihood estimates of
sigma and the acceleration model parameters are the values that maximize this
Likelihood. In most cases, these values are obtained by setting partial derivatives of the
log Likelihood to zero and solving the resulting (non-linear) set of equations.

The method As you can see, the procedure is complicated and computationally intensive, and only
is practical if appropriate software is available. It does have many desirable features such
complicated as:
and ● the method can, in theory at least, be used for any distribution model and
requires acceleration model and type of censored data
specialized
● estimates have "optimal" statistical properties as sample sizes (i.e. numbers of
software
failures) get large
● approximate confidence bounds can be calculated
● statistical tests of key assumptions can be made using the likelihood ratio test.
Some common tests are:
❍ the life distribution model vs another simpler model with fewer parameters
(i.e. a 3-parameter Weibull versus a 2-parameter Weibull, or a 2-parameter
Weibull vs an exponential)
❍ the constant slope from cell to cell consequence of typical acceleration
models
❍ the fit of a particular acceleration model
In general, the recommendations made when comparing methods of estimating life

distribution model parameters also apply here. Software incorporating acceleration
model analysis capability, while rare just a few years ago, is becoming more readily
available and many companies and universities have developed their own proprietary
versions.
Example Comparing Graphical Estimates and MLE 's
Arrhenius The data from the 3 stress cell Arrhenius example given in the preceding section were
example analyzed using a proprietary MLE program that could fit individual cells and also do an
comparing overall Arrhenius fit. The tables below compare results.
graphical
and MLE
method
Graphical Estimates MLE's
results
ln T50 Sigma ln T50 Sigma
Cell 1 8.17 .91 8.89 1.21

Cell 2 6.42 .66 6.47 .71
Cell 3 5.32 .81 5.33 .81
Acceleration Model Overall Estimates
H Sigma ln A
Graphical .808 .74 -18.312

MLE .863 .77 -19.91
Note that when there were a lot of failures and little censoring, the two methods are in
fairly close agreement. Both methods were also in close agreement on the Arrhenius
model results. However, even small differences can be important when projecting
reliability numbers at use conditions. In this example, the CDF at 25°C and 100,000
hours projects to .014 using the graphical estimates, and only .003 using the MLE
estimates.

MLE The Maximum Likelihood program also tested whether parallel lines (a single sigma)
method tests was reasonable, and whether the Arrhenius model was acceptable. The three cells of
models and data passed both of these Likelihood Ratio tests easily. In addition the MLE program
gives output included confidence intervals for all estimated parameters.
confidence
intervals SAS JMP™ software (previously used to find single cell Weibull MLE's) can also be
used for fitting acceleration models. This is shown next.
Using SAS JMP™Software To Fit Reliability Models
Detailed If you have JMP on your computer, set up to run as a browser application, click here to
explanation load a lognormal template JMP spreadsheet named arrex.jmp. This template has the
of how to Arrhenius example data already entered. The template extends JMP's analysis
use JMP capabilities beyond the standard JMP routines by making use of JMP's powerful
software to "Nonlinear Fit" option (links to blank templates for both Weibull and lognormal data
fit an are provided at the end of this page).
Arrhenius
model First a standard JMP reliability model analysis for this data will be shown. By working
with windows showing JMP and the Handbook, you can try out the steps in this
analysis as you read them.
The first part of the spread sheet should look as illustrated below.
Steps For Fitting The Arrhenius Model Using JMP's "Survival" Options
1. The "Start Time" column has all the fail and censor times and "Censor" and "Freq"
were entered as shown previously. In addition, the temperatures in degrees C
corresponding to each row were entered in "Temp in C". That is all that has to be
entered on the template; all other columns are calculated as needed. In particular, the
"1/kT" column contains the standard Arrhenius 1/kT values for the different temperature
cells.
2. To get a plot of all three cells, along with individual cell lognormal parameter
estimates, choose "Kaplan - Meier" (or "Product Limit") from the "Analysis" menu and
fill in the screen as shown below.

Column names are transferred to the slots on the right by highlighting them and clicking
on the tab for the slot. Note that the "Temp in C" column is transferred to the
"Grouping" slot in order to analyze and plot each of the three temperature cells
separately.
Clicking "OK" brings up the analysis screen below. All plots and estimates are based on
individual cell data, without the Arrhenius model assumption. Note: to obtain the
lognormal plots and parameter estimates and confidence bounds, it was necessary to
click on various "tabs" or "check" marks - this may depend on the software release
level.

This screen does not give -LogLikelihood values for the cells. These are obtained from
the "Parametric Model" option in the "Survival" menu (after clicking "Analyze").
3. First we will use the "Parametric Model" option to obtain individual cell estimates.
On the JMP data spreadsheet (arrex.jmp), select all rows except those corresponding to
cell 1 (the 85 degree cell) and choose "Exclude" from the "Row" button options (or do
"ctrl+E). Then click "Analyze" followed by "Survival" and "Parametric Model". Enter
the appropriate columns, as shown below, make sure you use "Get Model" to select
"lognormal" and click "Run Model".

This will generate a model fit screen for cell 1. Repeat for cells 2 and three. The three
resulting model fit screens are shown below.

Note that the model estimates and bounds are the same as obtained in step 2, but these
screens also give -LogLikelihood values. Unfortunately, as previously noted, these
values are off by the sum of the {ln times of failure} for each cell. These sums for the
three cells are 31.7871, 213.3097 and 371.2155, respectively. So the correct cell
-LogLikelihood values for comparing with other MLE programs are 53.3546, 265.2323
and 156.5250, respectively. Adding them together yields a total -LogLikelihood of
475.1119 for all the data fit with separate lognormal parameters for each cell (no
Arrhenius model assumption).
4. To fit the Arrhenius model across the three cells go back to the survival model
screen, this time with all the data rows included and the "1/kT" column selected and put
into the "Effects in Model" box via the "Add" button. This adds the Arrhenius
temperature effect to the MLE analysis of all the cell data. The screen looks like:
Clicking "Run Model" produces

The MLE estimates agree with those shown in the tables earlier on this page. The
-LogLikelihood for the model is given under "Full" in the output screen (and should be
adjusted by adding the sum of all the ln failure times from all three cells if comparisons
to other programs might be made). This yields a model -LogLikelihood of 105.4934 +
371.2155 = 476.7089.
5. The likelihood ratio test statistic for the Arrhenius model fit (which also incorporates
the single sigma acceleration assumption) is - 2Log , where is the difference
between the LogLikelihoods with and without the Arrhenius model assumption. Using
the results from steps 3 and 4, we have - 2Log = 2 × (476.709 - 475.112) = 3.194.
The degrees of freedom for the Chi Square test statistic are 6 - 3 = 3, since six
parameters were reduced to three under the acceleration model assumption. The chance
of getting a value 3.194 or higher is 36.3% for a Chi Square distribution with 3 D.O.F.,
which indicates an acceptable model (no significant lack of fit).
This completes a JMP Arrhenius model analysis of the three cells of data. Since the
Survival Modeling screen allows any "effects" to be included in the model, if different
cells of data had different voltages, the "ln V" column could be added as an effect to fit
the Inverse Power Law voltage model. In fact, several effects can be included at once, if
more than one stress varies across cells. Cross product stress terms could also be
included by adding these columns to the spreadsheet and adding them in the model as
additional "effects".

Arrhenius Steps For Fitting The Arrhenius Model Using the "Nonlinear Fit" Option and
example Special JMP Templates
using
special JMP There is another powerful and flexible tool included within JMP that can use MLE
template methods to fit reliability models. While this method requires some simple programming
and of JMP calculator equations, it offers the advantage of extending JMP's analysis
"Nonlinear capabilities to readout data (or truncated data, or any combination of different types of
Fit" data). Templates (available below) have been set up to cover lognormal and Weibull
data. The spreadsheet used above (arrex.jmp) is just the lognormal template, with the
Arrhenius data entered.
The following steps work with arrex.jmp because the "loss" columns have been set up
to calculate -LogLikelihoods for each row.
1. Load the arrex.jmp spreadsheet and Click "Analyze" on the Tool Bar and choose
"Nonlinear Fit".
2. Select the Loss (w/Temp) column and click "Loss" to put "Loss (w/Temp)" in the
box. This column on the spreadsheet automatically calculates the - LogLikelihood
values at each data point for the Arrhenius/lognormal model. Click "OK" to run the
Nonlinear Analysis.
3. You will next get a "Nonlinear Fit" screen. Select "Loss is -LogLikelihood" and click
the "Reset" and "Go" buttons to make sure you get a new analysis. The parameter
values for the constant Ln A (labeled "Con"), ∆H and sig will appear and the value of -
LogLikelihood is given under the heading "SSE". These numbers are -19.91, 0.863,
0.77 and 476.709, respectively. You can now click on "Confid Limits" to get upper and
lower confidence limits for these parameters. The stated value of "Alpha = .05" means
that the interval between the limits is a 95% confidence interval. At this point your
"Nonlinear Fit" screen looks as follows

4. Next you can run each cell separately by excluding all data rows corresponding to
other cells and repeating steps 1 through 3. For this analysis, select the "Loss (w/o
Stress)" column to put in "Loss" in step 2, since a single cell fit does not use
temperature . The numbers should match the table shown earlier on this page. The three
cell -LogLikelihood values are 53.355, 265.232 and 156.525. These add to 475.112,
which is the minimum -log likelihood possible, since it uses 2 independent parameters
to fit each cell separately (for a total of six parameters, overall).
The likelihood ratio test statistic for the Arrhenius model fit (which also incorporates
the single sigma acceleration assumption) is - 2Log λ = 2 x (476.709 - 475.112) =
3.194. Degrees of freedom for the Chi Square test statistic are 6 - 3 = 3, since six
parameters were reduced to three under the acceleration model assumption. The chance
of getting a value 3.194 or higher is 36.3% for a Chi Square distribution with 3 D.O.F.,
which indicates an acceptable model (no significant lack of fit).
For further examples of JMP reliability analysis there is an excellent collection of JMP
statistical tutorials put together by Professor Ramon Leon and one of his students, Barry
Eggleston, available on the Web at

http://www.nist.gov/cgi-bin/exit_nist.cgi?url=http://web.utk.edu/~leon/jmp/.
Data entry How To Use JMP Templates For Lognormal or Weibull Data (Including
on JMP Acceleration Model Analysis)
templates
for general With JMP installed to run as a browser application, you can click on weibtmp.jmp or
reliability lognmtmp.jmp and load (and save for later use) blank templates similar to the one
data shown above, for either Weibull or lognormal data data analysis. Here's how to enter
any kind of data on either of the templates.
Typical Data Entry
1. Any kind of censored or truncated or readout data can be entered. The rules are as
follows for the common case of (right) censored reliability data:
i) Enter exact failure times in the "Start Time" column, with "0" in the
"Cens" column and the number of failures at that exact time in the "Freq"
column.
ii) Enter temperature in degrees Celsius for the row entry in "Temp in C",
whenever data from several different operating temperatures are present
and an Arrhenius model fit is desired.
iii) Enter voltages in "Volt" for each row entry whenever data from several
different voltages are present and an Inverse Power Law model fit is
desired. If both temperatures and voltages are entered for all data rows, a
combined two stress model can be fit.
iv) Put censor times (where an unfailed units are removed from test, or no
longer observed) in both the "Start Time" and "Stop Time" column, and
enter "1" in the "Cens" column. Put the number of censored units in the
"Freq" column.
v) If readout (also known as interval) data is present, put the interval start
time and stop time in the corresponding columns and "2" in the "Cens"
column. Put the number of failures during the interval in the "Freq"
column. If the number of failures is zero, it doesn't matter if you include
the interval, or not.
Using The Templates For Model Fitting
Pick the appropriate template; weibtmp.jmp for a Weibull fit, or lognmtmp.jmp for a
lognormal fit.
After all the data have been entered, pick the appropriate "Loss" column and use the
Non Linear Fit analysis option (steps 1 - 4 above).
The "Loss (w/o Stress)" column is appropriate when only a straight life
distribution fit is desired, with no acceleration model.
The "Loss (w/Temp)" column handles an Arrhenius model (no other stress
taken into account). The "Loss (w/volts)" column handles an Inverse
Voltage Power Law model (no other stress taken into account).
The "Loss (Temp/Volt)" column handles a combined Arrhenius/Inverse
Voltage Power Law model fit.

All analyses are done the same as the 4 step "Non Linear Fit" example shown above.
A few tricks are needed to handle the rare cases of truncated data or left censored data -
these are described next.
How to JMP Template Data Entry For Truncated or Left Censored Weibull or
handle Lognormal Data
truncated or
left Left censored data means all exact times of failure below a lower cut off time T0 are
censored unknown, but the number of these failures is known. Merely enter an interval with start
data using time 0 and stop time T0 on the appropriate template and put "2" in the "Cens" column
JMP and the number in the "Freq" column.
templates
Left truncated data means all data points below a lower cut off point T0 are unknown,
and even the number of such points is unknown. This situation occurs commonly for
measurement data, when the measuring instrument has a lower threshold detection limit
at T0. Assume there are n data points (all above T0) actually observed. Enter the n
points as you normally would on the appropriate template ("Cens" gets 0 and "Freq"
gets 1) and add a start time of T0 with a "Cens" value of 1 and a "Freq" value of -n (yes,
minus n!).
Right truncated data means all data points above an upper cut off point T1 are
unknown, and even the number of such points is unknown. Assume there are n data
points (all below T1) actually observed. Enter the n points as you normally would on the
appropriate template ("Cens" gets 0 and "Freq" gets 1) and add a start time of 0 and a
stop time of T1 with a "Cens" value of 2 and a "Freq" value of -n (yes, minus n!)

8.4.2.3. Fitting models using degradation data instead of failures

8.4.2.3. Fitting models using degradation data instead

of failures
If you can fit When failure can be related directly to a change over time in a measurable product
models using parameter, it opens up the possibility of measuring degradation over time and using that data
degradation to extrapolate when failure will occur. That allows us to fit acceleration models and life
data, you distribution models without actually waiting for failures to occur.
don't need
actual test This overview of degradation modeling assumes you have a chosen life distribution model
failures and an acceleration model and offers an alternative to the accelerated testing methodology
based on failure data, previously described. The following topics are covered:
● Common assumptions
● Advantages
● Drawbacks
● A simple method
● A more accurate approach for a special case
● Example
More details can be found in Nelson (1990, pages 521-544) or Tobias and Trindade (1995,
pages 197-203).
Common Assumptions When Modeling Degradation Data
You need a Two common assumptions typically made when degradation data are modeled are the
measurable following:
parameter 1. A parameter D, that can be measured over time, drifts monotonically (upwards, or
that drifts downwards) towards a specified critical value DF. When it reaches DF, failure occurs.
(degrades)
2. The drift, measured in terms of D, is linear over time with a slope (or rate of
linearly to a
degradation) R, that depends on the relevant stress the unit is operating under and also
critical
the (random) characteristics of the unit being measured. Note: it may be necessary to
failure value
define D as a transformation of some standard parameter in order to obtain linearity -
logarithms or powers are sometimes needed.
The figure below illustrates these assumptions by showing degradation plots of 5 units on
test. Degradation readings for each unit are taken at the same four time points and straight
lines fit through these readings on a unit by unit basis. These lines are then extended up to a
critical (failure) degradation value. The projected times of failure for these units are then read
off the plot. The are: t1, t2, ...,t5.

Plot of
linear
degradation
trends for 5
units read
out at four
time points
In many practical situations, D starts at 0 at time zero, and all the linear theoretical
degradation lines start at the origin. This is the case when D is a "% change" parameter, or
failure is defined as a change of a specified magnitude in a parameter, regardless of its
starting value. Lines all starting at the origin simplify the analysis, since we don't have to
characterize the population starting value for D, and the "distance" any unit "travels" to reach
failure is always the constant DF. For these situations, the degradation lines would look as
follows:
Often, the
degradation
lines go
through the
origin - as
when %
change is the
measurable
parameter
increasing to
a failure
level

It is also common to assume the effect of measurement error, when reading values of D, has
relatively little impact on the accuracy of model estimates.
Advantages of Modeling Based on Degradation Data
Modeling 1. Every degradation readout for every test unit contributes a data point. This leads to
based on large amounts of useful data, even if there are very few failures.
complete 2. You don't have to run tests long enough to obtain significant numbers of failures.
samples of
3. You can run low stress cells that are much closer to use conditions and obtain
measurement
meaningful degradation data. The same cells would be a waste of time to run if failures
data, even
were needed for modeling. Since these cells are more typical of use conditions, it
with low
makes sense to have them influence model parameters.
stress cells,
offers many 4. Simple plots of degradation vs time can be used to visually test the linear degradation
advantages assumption.
Drawbacks to Modeling Based on Degradation Data
Degradation 1. For many failure mechanisms, it is difficult or impossible to find a measurable

may not parameter that degrades to a critical value in such a way that reaching that critical
proceed in a value is equivalent to what the customer calls a failure.
smooth, 2. Degradation trends may vary erratically from unit to unit, with no apparent way to
linear transform them into linear trends.
fashion
3. Sometimes degradation trends are reversible and a few units appear to "heal
towards
themselves" or get better. This kind of behavior does not follow typical assumptions
what the
and is difficult to model.
customer
calls 4. Measurement error may be significant, and overwhelm small degradation trends,
"failure" especially at low stresses.
5. Even when degradation trends behave according to assumptions and the chosen
models fit well, the final results may not be consistent with an analysis based on actual

failure data. This probably means that the failure mechanism depends on more than a
simple continuous degradation process.
Because of the last listed drawback, it is a good idea to have at least one high stress cell
where enough real failures occur to do a standard life distribution model analysis. The
parameters obtained can be compared to the predictions from the degradation data analysis,
as a "reality" check.
A Simple Method For Modeling Degradation Data
A simple 1. As shown in the figures above, fit a line through each unit's degradation readings. This
approach is can be done by hand, but using a least squares regression program is better (like
to extend Dataplot's "LINEAR FIT Y X" or EXCEL's line fitting routines).
each units 2. Take the equation of the fitted line, substitute DF for Y and solve for X. This value of
degradation X is the "projected time of fail" for that unit.
line until a
3. Repeat for every unit in a stress cell until a complete sample of (projected) times of
projected
failure is obtained for the cell.
"failure
time" is 4. Use the failure times to compute life distribution parameters for a cell. Under the fairly
obtained typical assumption of a lognormal model, this is very simple. Take natural logarithms
of all failure times and treat the resulting data as a sample from a normal distribution.
Compute the sample mean and the sample standard deviation. These are estimates of
ln T50 and , respectively, for the cell.
5. Assuming there are k cells with varying stress, fit an appropriate acceleration model
using the cell ln T50's, as described in the graphical estimation section. A single sigma
estimate is obtained by taking the square root of the average of the cell estimates
(assuming the same number of units each cell). If the cells have nj units on test, where
the nj's are not all equal, use the pooled sum of squares estimate across all k cells
calculated by
A More Accurate Regression Approach For the Case Where D = 0 at time 0 and the
"Distance To Fail" DF is the Same for All Units
Models can Let the degradation measurement for the i-th unit at the j-th readout time in the k-th stress
be fit using cell be given by Dijk, and let the corresponding readout time for this readout be denoted by tjk
all the . That readout gives a degradation rate (or slope) estimate of Dijk / tjk. This follows from the
degradation linear assumption or:
readings and
linear (Rate of degradation) × (Time on test) = (Amount of degradation)
regression
Based on that readout alone, an estimate of the natural logarithm of the time to fail for that
unit is
yijk = ln DF - (ln Dijk - ln tjk).
This follows from the basic formula connecting linear degradation with failure time
(rate of degradation) × (time of failure) = DF
by solving for (time of failure) and taking natural logarithms.

For an Arrhenius model analysis, where
where the xk values are 1/KT. Here T is the temperature of the k-th cell, measured in Kelvin
(273.16 + degrees Celsius) and K is Boltzmann's constant (8.617 × 10-5 in eV/ unit Kelvin).
Use a linear regression program to estimate a = ln A and b = h. If we further assume tf has
a lognormal distribution, the mean square residual error from the regression fit is an
estimate of (where is the lognormal sigma).

One way to think about this model is as follows: each unit has a random rate R of
degradation. Since tf = DF/R, it follows that if tf is lognormal, then R must also have a
lognormal distribution. After we take logarithms, ln R has a normal distribution with a mean
determined by the acceleration model parameters. The randomness in R comes from the
variability in physical characteristics from unit to unit, due to material and processing
differences.
Note: the sigma estimate for this method and the simpler graphical approach might tend to
overestimate slightly, because it includes an adder due to the measurement error that occurs
when making degradation readouts. This is generally assumed to have an insignificant
impact.
Example: Arrhenius Degradation Analysis
An example A component has a critical parameter that studies show degrades linearly over time at a rate
using the that varies with operating temperature. A component failure based on this parameter occurs
regression when the parameter value changes by 30% or more. Fifteen components were tested under 3
approach to different temperature conditions (5 at 65o, 5 at 85o and the last 5 at 105o). Degradation
fit an percent values were read out at 200, 500 and 1000 hours. The readings are given by unit in
Arrhenius the following three temperature cell tables.
model
65 Degrees C
200 hr 500 hr 1000 hr
Unit 1 .87 1.48 2.81
Unit 2 .33 .96 2.13
Unit 3 .94 2.91 5.67
Unit 4 .72 1.98 4.28
Unit 5 .66 .99 2.14
85 Degrees C
200 hr 500 hr 1000 hr
Unit 1 1.41 2.47 5.71
Unit 2 3.61 8.99 17.69
Unit 3 2.13 5.72 11.54
Unit 4 4.36 9.82 19.55
Unit 5 6.91 17.37 34.84

105 Degrees C
200 hr 500 hr 1000 hr
Unit 1 24.58 62.02 124.10
Unit 2 9.73 24.07 48.06
Unit 3 4.74 11.53 23.72
Unit 4 23.61 58.21 117.20
Unit 5 10.90 27.85 54.97
Note that one unit failed in the 85 degree cell and 4 units failed in the 105 degree cell.
Because there were so few failures, it would be impossible to fit a life distribution model in
any cell but the 105 degree cell, and therefore no acceleration model can be fit using failure
data. We will fit an Arrhenius/Lognormal model, using the degradation data.
Dataplot Solution:
Dataplot From the above tables, first create a data row of 45 degradation values starting with the first
easily fits the row in the first table and proceeding to the last row in the last table. Put these in a text file
model to the called DEGDAT. DEGDAT has one row of 45 numbers looking like the following: .87, .33,
degradation .99, .72, .66, 1.28, .96, 2.91, 1.98, .99, . . . , 124.10, 48.06, 23.72, 117.20, 54.97.
data
Next, create a text file TEMPDAT, containing the corresponding 45 temperatures. TEMP has
Other 15 repetitions of 65, followed by 15 repetitions of 85 and then 15 repetitions of 105.
regression
programs Finally, create a text file TIMEDAT, containing the corresponding readout times. These are
would work 200, 200, 200, 200, 200, 500, 500, 500, 500, 500, 1000, 1000, 1000, 1000, repeated 3 times.
equally well Assuming the data files just created are placed in the Dataplot directory, the following
commands will complete the analysis:.
READ DEGDAT. DEG
READ TEMPDAT. TEMP READ TIMEDAT. TIME LET YIJK = LOG(30) -
(LOG(DEG) - LOG(TIME)) LET XIJK = 100000/(8.617*(TEMP + 273.16))
LINEAR FIT YIJK XIJK
The output is (with unnecessary items edited out)
LEAST SQUARES POLYNOMIAL FIT
SAMPLE SIZE N = 45
DEGREE = 1
PARAMETER ESTIMATES (APPROX ST. DEV) tVALUE
1 A0 -18.9434 (1.833) -10
2 A1 .818774 (.5641e-01) 15
RESIDUAL STANDARD DEVIATION = .5610512000

The Arrhenius model parameter estimates are: ln A = -18.94; H = .82. An estimate for the
lognormal sigma is = .56.

8.4.3. How do you project reliability at use conditions?

8.4.3. How do you project reliability at use

conditions?
When General Considerations
projecting
from high Reliability projections based on failure data from high stress tests are
stress to use based on assuming we know the correct acceleration model for the
conditions, failure mechanism under investigation and we are also using the correct
having a life distribution model. This is because we are extrapolating
correct "backwards" - trying to describe failure behavior in the early tail of the
acceleration life distribution, where we have little or no actual data.
model and For example, with an acceleration factor of 5000 (and some are much
life larger than this) the first 100,000 hours of use life is "over" by 20 hours
distribution into the test. Most, or all, of the test failures typically come later in time
model is and are used to fit a life distribution model where only the first 20 hours
critical or less are of practical use. Many distributions may be flexible enough
to adequately fit the data at the percentiles where the points are, and yet
differ by orders of magnitude in the very early percentiles (sometimes
referred to as the early "tail" of the distribution).
However, it is frequently necessary to test at high stress (to get any
failures at all!) and project backwards to use. When doing this bear in
mind two important points:

Project for ● Distribution models, and especially acceleration models, should

each failure be applied only to a single failure mechanism at a time. Separate
mechanism out failure mechanisms when doing the data analysis and use the
separately competing risk model to build up to a total component failure rate
Try to find theoretical justification for the chosen models, or at
●
least a successful history of their use for the same or very similar
mechanisms. (Choosing models solely based on empirical fit is
like extrapolating from quicksand to a mirage.)
How to Project from High Stress to Use Stress
Two types of use condition reliability projections are common:
1. Projection to use conditions after completing a multiple stress cell
experiment and successfully fitting both a life distribution model
and an acceleration model
2. Projection to use conditions after a single cell at high stress is run
as a line reliability monitor
Arrhenius The Arrhenius example from the graphical estimation and the MLE
model estimation sections ended by comparing use projections of the CDF at
projection 100,000 hours. This is a projection of the first type. We know from the
example Arrhenius model assumption that the T50 at 25°C is just
with
Dataplot
commands
Using the graphical model estimates for ln A and we have
T50 at use = e-18.312 × e.808 × 11605/298.16 = e13.137 = 507380
and combining this T50 with the common sigma of .74 allows us to
easily estimate the CDF or failure rate after any number of hours of
operation at use conditions. In particular, the Dataplot command
LET Y = LGNCDF((T/T50),sigma)
evaluates a lognormal CDF at time T, and
LET Y = LGNCDF((100000/507380),.74)
returns the answer .014 given in the MLE estimation section as the
graphical projection of the CDF at 100,000 hours at a use temperature of
25°C.
If the life distribution model had been Weibull, the same type of
analysis would be done by letting the characteristic life parameter
vary with stress according to the acceleration model, while the shape

parameterγ is constant for all stress conditions.

The second type of use projection was used in the section on lognormal
and Weibull Tests where we judged new lots of product by looking at
the proportion of failures in a sample tested at high stress. The
assumptions we made were:
● we knew the acceleration factor between use and high stress
● the shape parameter (sigma for the lognormal, gamma for the
Weibull) is also known and does not change significantly from lot
to lot.
With these assumptions, we can take any proportion fails we see from a
high stress test and project a use CDF or failure rate. For a T hour high
stress test and an acceleration factor of A from high stress to use stress,
an observed proportion p is converted to a use CDF at 100,000 hours for
a lognormal model as follows:
LET T50STRESS = T *LGNPPF(p, )
LET CDF = LGNCDF((100000/(A*T50STRESS)), )
If the model is Weibull, the Dataplot commands are
LET ASTRESS = T*WEIPPF(p, )
LET CDF = WEICDF((100000/(A*ASTRESS)), )

8.4.4. How do you compare reliability between two or more populations?

8.4.4. How do you compare reliability

between two or more populations?
Several Comparing reliability among populations based on samples of failure
methods for data usually means asking whether the samples came from populations
comparing with the same reliability function (or CDF). Three techniques already
reliability described can be used to answer this question for censored reliability
between data. These were:
populations
● Comparing sample proportion fails
are
described ● Likelihood ratio test comparisons
● Lifetime regression comparisons
Comparing Sample Proportion Fails
Assume each sample is a random sample from possibly a different lot,
vendor or production plant. All the samples are tested under the same
conditions. Each has an observed proportion of failures on test. Call
these sample proportion of failures p1, p2, p3, ...pn. Could these all have
come from equivalent populations?
This is a question covered in Chapter 7, and the techniques described
there apply equally well here.
Likelihood Ratio Test Comparisons
The Likelihood Ratio test was described earlier. In this application, the
Likelihood ratio has as a denominator the product of all the
Likelihoods of all the samples assuming each population has its own
unique set of parameters. The numerator is the product of the
Likelihoods assuming the parameters are exactly the same for each
population. The test looks at whether -2ln is unusually large, in
which case it is unlikely the populations have the same parameters (or
reliability functions).
This procedure is very effective if, and only if, it is built into the

8.4.4. How do you compare reliability between two or more populations?
analysis software package being used and this software covers the
models and situations of interest to the analyst.
Lifetime Regression Comparisons
Lifetime regression is similar to maximum likelihood and likelihood
ratio methods. Each sample is assumed to come from a population with
the same shape parameter and a wide range of questions about the scale
parameter (which is often assumed to be a "measure" of lot to lot or
vendor to vendor quality) can be formulated and tested for significance.
For a complicated, but realistic example, assume a company
manufactures memory chips and can use chips with some known defects
("partial goods") in many applications. However, there is a question of
whether the reliability of "partial good" chips is equivalent to "all good"
chips. There exists lots of customer reliability data to answer this
question - however the data is difficult to analyze because it contains
several different vintages with known reliability differences as well as
chips manufactured at many different locations. How can the partial
good vs all good question be resolved?
A lifetime regression model can be set up with variables included that
change the scale parameter based on vintage, location , partial vs all
good, and any other relevant variables. Then, a good lifetime regression
program will sort out which, if any, of these factors are significant and,
in particular, whether there is a significant difference between "partial
good" and "all good".
Software that will do lifetime regression is not widely available at this
time, however.

8.4.5. How do you fit system repair rate models?

8.4.5. How do you fit system repair rate

models?
Fitting This sub-section describes how to fit system repair rate models when
models you have actual failure data. The data could come from from observing
discussed a system in normal operation or from running tests such as Reliability
earlier Improvement tests.
The three models covered are the constant repair rate (HPP/exponential)
model, the power law (Duane) model and the exponential law model.

8.4.5.1. Constant repair rate (HPP/exponential) model

8.4.5.1. Constant repair rate

(HPP/exponential) model
This The HPP or exponential model is widely used for two reasons:
section ● Most systems spend most of their useful lifetimes operating in the
covers flat constant repair rate portion of the bathtub curve
estimating
MTBF's ●It is easy to plan tests, estimate the MTBF and calculate
and confidence intervals when assuming the exponential model
calculating This section covers the following:
upper and 1. Estimating the MTBF (or repair rate/failure rate)
lower
confidence 2. How to use the MTBF confidence interval Factors
bounds 3. Tables of MTBF confidence interval factors
4. Confidence interval equation and "zero fails" case
5. Dataplot/EXCEL calculation of confidence Intervals
6. Example
Estimating the MTBF (or repair rate/failure rate)
For the HPP system model, as well as for the non repairable exponential
population model, there is only one unknown parameter (or
equivalently, the MTBF = 1/ ) . The method used for estimation is the
same for the HPP model and for the exponential population model.

The best The estimate of the MTBF is

estimate of
the MTBF
is just
"Total
Time"
divided by
"Total
Fails"
This estimate is the maximum likelihood estimate whether the data is

censored or complete, or from a repairable system or a non repairable
population.
Confidence How To Use the MTBF Confidence Interval Factors

Interval 1. Estimate the MTBF by the standard estimate (total unit test hours
Factors divided by total fails)
multiply the
estimated 2. Pick a confidence level (i.e. pick 100x(1- )). For 95%, = .05;
MTBF to for 90%, = .1; for 80%, = .2 and for 60%, = .4
obtain 3. Read off a lower and an upper factor from the confidence interval
lower and tables for the given confidence level and number of failures r
upper 4. Multiply the MTBF estimate by the lower and upper factors to
bounds on obtain MTBFlower and MTBFupper
the true
MTBF 5. When r (the number of failures) = 0, multiply the total unit test
hours by the "0 row" lower factor to obtain a 100 × (1- /2)
one-sided lower bound for the MTBF. There is no upper bound
when r = 0.
6. Use (MTBFlower, MTBFupper) as a 100 × (1- confidence interval
for the MTBF (r>0)
7. Use MTBFlower as a (one-sided) lower 100×(1- /2) limit for the
MTBF
8. Use MTBFupper as a (one-sided) upper 100x(1- /2) limit for the
MTBF
9. Use (1/MTBFupper, 1/MTBFlower) as a 100×(1- ) confidence
interval for
10. Use 1/MTBFupper as a (one-sided) lower 100×(1- /2) limit for

11. Use 1/MTBFlower as a (one-sided) upper 100×(1- /2) limit for

Tables of MTBF Confidence Interval Factors
Confidence Confidence Interval Factors to Multiply MTBF Estimate

bound
factor 60% 80%
tables for Num Lower for Upper for Lower for Upper for
60, 80, 90 Fails r MTBF MTBF MTBF MTBF
and 95%
confidence 0 0.6213 - 0.4343 -
1 0.3340 4.4814 0.2571 9.4912
2 0.4674 2.4260 0.3758 3.7607
3 0.5440 1.9543 0.4490 2.7222
4 0.5952 1.7416 0.5004 2.2926
5 0.6324 1.6184 0.5391 2.0554
6 0.6611 1.5370 0.5697 1.9036
7 0.6841 1.4788 0.5947 1.7974
8 0.7030 1.4347 0.6156 1.7182
9 0.7189 1.4000 0.6335 1.6567
10 0.7326 1.3719 0.6491 1.6074
11 0.7444 1.3485 0.6627 1.5668
12 0.7548 1.3288 0.6749 1.5327
13 0.7641 1.3118 0.6857 1.5036
14 0.7724 1.2970 0.6955 1.4784
15 0.7799 1.2840 0.7045 1.4564
20 0.8088 1.2367 0.7395 1.3769
25 0.8288 1.2063 0.7643 1.3267
30 0.8436 1.1848 0.7830 1.2915
35 0.8552 1.1687 0.7978 1.2652
40 0.8645 1.1560 0.8099 1.2446
45 0.8722 1.1456 0.8200 1.2280
50 0.8788 1.1371 0.8286 1.2142
75 0.9012 1.1090 0.8585 1.1694
100 0.9145 1.0929 0.8766 1.1439
500 0.9614 1.0401 0.9436 1.0603
Confidence Interval Factors to Multiply MTBF Estimate
90% 95%
Num Lower for Upper for Lower for Upper for
Fails MTBF MTBF MTBF MTBF
0 0.3338 - 0.2711 -

1 0.2108 19.4958 0.1795 39.4978

2 0.3177 5.6281 0.2768 8.2573
3 0.3869 3.6689 0.3422 4.8491
4 0.4370 2.9276 0.3906 3.6702
5 0.4756 2.5379 0.4285 3.0798
6 0.5067 2.2962 0.4594 2.7249
7 0.5324 2.1307 0.4853 2.4872
8 0.5542 2.0096 0.5075 2.3163
9 0.5731 1.9168 0.5268 2.1869
10 0.5895 1.8432 0.5438 2.0853
11 0.6041 1.7831 0.5589 2.0032
12 0.6172 1.7330 0.5725 1.9353
13 0.6290 1.6906 0.5848 1.8781
14 0.6397 1.6541 0.5960 1.8291
15 0.6494 1.6223 0.6063 1.7867
20 0.6882 1.5089 0.6475 1.6371
25 0.7160 1.4383 0.6774 1.5452
30 0.7373 1.3893 0.7005 1.4822
35 0.7542 1.3529 0.7190 1.4357
40 0.7682 1.3247 0.7344 1.3997
45 0.7800 1.3020 0.7473 1.3710
50 0.7901 1.2832 0.7585 1.3473
75 0.8252 1.2226 0.7978 1.2714
100 0.8469 1.1885 0.8222 1.2290
500 0.9287 1.0781 0.9161 1.0938
Confidence Interval Equation and "Zero Fails" Case
Formulas Confidence bounds for the typical Type I censoring situation are obtained
for from chi-square distribution tables or programs. The formula for
confidence calculating confidence intervals is:
bound
factors -
even for
"zero fails"
case
In this formula, is a value that the chi-square distribution

with 2r degrees of freedom is greater than with probability 1- /2. In
other words, the right hand tail of the distribution has probability 1- /2.
A even simpler version of this formula can be written using T = the total
unit test time:

These bounds are exact for the case of one or more repairable systems on
test for a fixed time. They are also exact when non repairable units are on
test for a fixed time, and failures are replaced with new units during the
course of the test. For other situations, they are approximate.
When there are zero failures during the test or operation time, only a
(one-sided) MTBF lower bound exists, and this is given by
MTBFlower = T/-ln
The interpretation of this bound is the following: if the true MTBF were
any lower than MTBFlower, we would have seen at least one failure
during T hours of test with probability at least 1- . Therefore, we are
100×(1- ) confident that the true MTBF is not lower than MTBFlower.
Dataplot/EXCEL Calculation of Confidence Intervals
Dataplot A lower 100×(1- /2) confidence bound for the MTBF is given by
and
EXCEL LET LOWER = T*2/CHSPPF([1- /2],[2*(r+1)])
calculation where T is the total unit or system test time and r is the total number of
of failures.
confidence
limits The upper 100×(1- /2) confidence bound is
LET UPPER = T*2/CHSPPF( /2,[2*r])
and (LOWER, UPPER) is a 100×(1- ) confidence interval for the true
MTBF.
The same calculations can be done with EXCEL built in functions with
the commands
=T*2/CHINV([ /2],[2*(r+1)]) for the lower bound and
=T*2/CHINV([1- /2],[2*r]) for the upper bound.
Note that the Dataplot CHSPPF function requires left tail probability
inputs (i.e. /2 for the lower bound and 1- /2 for the upper bound),
while the EXCEL CHINV function requires right tail inputs (i.e. 1- /2
for the lower bound and /2 for the upper bound).
Example

Example A system was observed for two calendar months of operation, during
showing which time it was in operation for 800 hours and it had 2 failures.
how to
calculate The MTBF estimate is 800/2 = 400 hours. A 90% confidence interval is
confidence given by (400×.3177, 400×5.6281) = (127, 2251). The same interval
limits could have been obtained using the Dataplot commands
LET LOWER = 1600/CHSPPF(.95,6)
LET UPPER = 1600/CHSPPF(.05,4)
or the EXCEL commands
=1600/CHINV(.05,6) for the lower limit
=1600/CHINV(.95,4) for the upper limit.
Note that 127 is a 95% lower limit for the true MTBF. The customer is
usually only concerned with the lower limit and one-sided lower limits
are often used for statements of contractual requirements.
Zero fails What could we have said if the system had had no failures? For a 95%
confidence lower confidence limit on the true MTBF, we either use the 0 fails factor
limit from the 90% confidence interval table and calculate 800 × .3338 = 267
calculation or we use T/-ln = 800/-ln .05 = 267.

8.4.5.2. Power law (Duane) model


The Power Brief Review of Power Law Model and Duane Plots
Law
(Duane) Recall that the Power Law is a NHPP where the expected number of fails M(t)
model has and the repair rate M'(t) = m(t), are given by:
been very
successful
for modeling
industrial
The parameter = 1-b is called the Reliability Growth Slope, and typical
reliability
improvement industry values for growth slopes during reliability improvement tests are in
data the .3 to .6 range.
If a system is observed for a fixed time of T hours, and failures occur at times
t1, t2, t3, ..., tr (where the start of the test or observation period is time 0), then a
Duane plot is a plot of (ti / i) versus ti on log log graph paper. If the data are
consistent with a Power Law model, the points in a Duane Plot will roughly
follow a straight line with slope and intercept (where t = 1 on the log log
paper) of -log10a.
MLE's for Estimates for the Power Law Model

the Power
Law model Computer aided graphical estimates can easily be obtained by doing a
are given regression fit of Y = ln (ti / i) vs X = ln ti. The slope is the estimate and
e-intercept is the a estimate. The estimate of b is 1- . The Dataplot command

for the regression fit is FIT Y X.
However, better estimates can easily be calculated. These are modified
maximum likelihood estimates (corrected to eliminate bias). The formulas are
given below for a fixed time of T hours, and r failures occurring at times t1, t2,
t3, ..., tr.

The estimated MTBF at the end of the test (or observation) period is
Approximate Approximate Confidence Bounds for the MTBF at End of Test

confidence
bounds for We give an approximate 100 × (1- )% confidence interval (ML, MU) for the
the MTBF at MTBF at end of test. Note that ML is a 100×(1- /2)% lower bound and MU is
end of test a 100×(1- /2)% upper bound. The formulas are:
are given
where is the upper 100×(1- /2) percentile point of the standard normal
distribution.

There is a Dataplot Macro for Estimates And Confidence Bounds

Dataplot
macro for The Dataplot macro Powerest.dp will calculate β, a, and the MTBF at the end
doing these of test, along with a 100x(1-α) percent confidence interval for the true MTBF
calculations at the end of test (assuming, of course, that the Power Law model holds).
Before calling up Powerest.dp, the user must define the following variables:
● T = the total test time
● R = the total number of failures
● TIMES = the actual system times of failure (the start of test is time 0)
An example of the use of Powerest.dp is given next.

Case Study 1: Reliability Improvement Test Data Continued
Dataplot This case study was introduced in section 2, where we did various plots of the
macro used data, including a Duane Plot. The case study was continued when we discussed
to fit Power trend tests and verified that significant improvement had taken place. Now we
Law to Case will use Powerest.dp to complete the case study data analysis.
Study 1 data
The observed failure times were: 5, 40, 43, 175, 389, 712, 747, 795, 1299 and
1478 hours, with the test ending at 1500 hours. The Dataplot session looks as
follows:
LET T = 1500
LET R = 10
LET TIMES = DATA 5 40 43 175 389 712 747 795
1299 1478
CALL POWEREST.DP
THIS PROGRAM ASSUMES YOU HAVE ALREADY

CREATED THE FOLLOWING VARIABLES
T = THE TOTAL SYSTEM TEST TIME

R = THE TOTAL NUMBER OF FAILURES (R
GREATER THAN 2)
TIMES = A DATA VECTOR OF THE R SYSTEM
FAILURE TIMES
DO YOU WANT TO CONTINUE?
IF NO, ENTER 0
IF YES, ENTER 1
1
INPUT YOUR ALPHA FOR 100 X (1-ALPHA) CONFIDENCE

INTERVAL
.2
THE RELIABILITY GROWTH SLOPE BETA IS 0.516495

THE A PARAMETER IS 0.2913
THE MTBF AT END OF TEST IS 310.234
THE DESIRED 80 PERCENT CONFIDENCE INTERVAL IS:

(157.7139 , 548.5565)
AND 157.7139 IS A (ONE-SIDED) 90 PERCENT
LOWER LIMIT

8.4.5.3. Exponential law model


Estimates Recall from section 1 that the Exponential Law refers to a NHPP
for the
parameters process with repair rate M'(t) = m(t) = . This model has not been
of the used nearly as much in industrial applications as the Power Law model,
Exponential and it is more difficult to analyze. Only a brief description will be given
Law model here.
can be
obtained Since the expected number of failures is given by
from either
a graphical M(t) = and ln M(t) = , a plot of the cum
procedure fails versus time of fail on log-linear paper should roughly follow a
or maximum straight line with slope . Doing a regression fit of y = ln cum fails
likelihood versus x = time of fail will provide estimates of the slope and the
estimation
intercept -ln .
Alternatively, maximum likelihood estimates can be obtained from the

following pair of equations:
The first equation is non-linear and must be solved iteratively to obtain

the estimate. Then, this estimate is substituted into the second
equation to solve for the estimate.


8.4.6. How do you estimate reliability using the Bayesian gamma prior model?

8.4.6. How do you estimate reliability using the

Bayesian gamma prior model?
The Bayesian paradigm was introduced in Section 1 and Section 2 described the assumptions
underlying the gamma/exponential system model (including several methods to transform prior
data and engineering judgment into gamma prior parameters "a" and "b"). Finally, we saw in
Section 3 how to use this Bayesian system model to calculate the required test time needed to
confirm a system MTBF at a given confidence level.
Review of The goal of Bayesian reliability procedures is to obtain as accurate a posterior distribution as
Bayesian possible, and then use this distribution to calculate failure rate (or MTBF) estimates with
procedure confidence intervals (called credibility intervals by Bayesians). The figure below summarizes
for the the steps in this process.
gamma
exponential
system
model

How to Once the test has been run, and r failures observed, the posterior gamma parameters are:
estimate a' = a + r, b' = b + T
the MTBF
and a (median) estimate for the MTBF, using EXCEL, is calculated by
with
bounds, = 1/GAMMAINV(.5, a', (1/ b'))
based on Some people prefer to use the reciprocal of the mean of the posterior distribution as their estimate
the
for the MTBF. The mean is the minimum mean square error (MSE) estimator of , but using
posterior
the reciprocal of the mean to estimate the MTBF is always more conservative than the "even
distribution
money" 50% estimator.
A lower 80% bound for the MTBF is obtained from
= 1/GAMMAINV(.8, a', (1/ b'))
and, in general, a lower 100×(1- )% lower bound is given by
= 1/GAMMAINV((1-), a', (1/ b')).
A two sided 100×(1- )% credibility interval for the MTBF is
[{= 1/GAMMAINV((1- /2), a', (1/ b'))},{= 1/GAMMAINV(( /2), a', (1/ b'))}].
Finally, = GAMMADIST((1/M), a', (1/b'), TRUE) calculates the probability the MTBF is greater
than M.
Example
A Bayesian A system has completed a reliability test aimed at confirming a 600 hour MTBF at an 80%
example confidence level. Before the test, a gamma prior with a = 2, b = 1400 was agreed upon, based on
using testing at the vendor's location. Bayesian test planning calculations, (allowing up to 2 new
EXCEL to failures) called for a test of 1909 hours. When that test was run, there actually were exactly two
estimate failures. What can be said about the system?
the MTBF
and The posterior gamma CDF has parameters a' = 4 and b' = 3309. The plot below shows CDF
calculate values on the y-axis, plotted against 1/ = MTBF, on the x-axis. By going from probability, on
upper and the y-axis, across to the curve and down to the MTBF, we can read off any MTBF percentile
lower point we want. (The EXCEL formulas above will give more accurate MTBF percentile values
bounds than can be read off a graph.)

The MTBF values are shown below:

= 1/GAMMAINV(.9, 4, (1/ 3309)) has value 495 hours
= 1/GAMMAINV(.8, 4, (1/ 3309)) has value 600 hours (as expected)
The test has confirmed a 600 hour MTBF at 80% confidence, a 495 hour MTBF at 90 %
confidence and (495, 1897) is a 90 percent credibility interval for the MTBF. A single number
(point) estimate for the system MTBF would be 901 hours. Alternatively, you might want to use
the reciprocal of the mean of the posterior distribution (b'/a') = 3309/4 = 827 hours as a single
estimate. The reciprocal mean is more conservative - in this case it is a 57% lower bound, as
=GAMMADIST((4/3309),4,(1/3309),TRUE) shows.

8.4.7. References For Chapter 8: Assessing ProductReliability

8.4.7. References For Chapter 8: Assessing

ProductReliability
Aitchison, J. and Brown, J. A. C.,(1957), The Log-normal distribution, Cambridge
University Press, New York and London.
Ascher, H. (1981), 'Weibull Distribution vs Weibull Process," Proceedings Annual
Reliability and Maintainability Symposium, pp. 426-431.
Ascher, H. and Feingold, H. (1984), Repairable Systems Reliability, Marcel Dekker,
Inc., New York.
Bain, L.J. and Englehardt, M. (1991), Statistical Analysis of Reliability and Life-Testing
Models: Theory and Methods, 2nd ed., Marcel Dekker, New York.
Barlow, R. E. and Proschan, F. (1975), Statistical Theory of Reliability and Life Testing,
Holt, Rinehart and Winston, New York
Birnbaum, Z.W., and Saunders, S.C. (1968), "A Probabilistic Interpretation of Miner's
Rule," SIAM Journal of Applied Mathematics, Vol. 16, pp. 637-652.
Birnbaum, Z.W., and Saunders, S.C. (1969), "A New Family of Life Distributions,"
Journal of Applied Probability, , Vol. 6, pp. 319-327.
Cox, D.R. and Lewis, P.A.W. (1966), The Statistical Analysis of Series of Events, John
Wiley & Sons, Inc., New York.
Cox, D.R. (1972), "Regression Models and Life Tables," Journal of the Royal Statistical,
B 34, pp. 187-220
Cox, D. R., and Oakes, D. (1984), Analysis of Survival Data, Chapman and Hall,
London, New York.
Crow, L.H. (1974), "Reliability Analysis for Complex Repairable Systems, "Reliability
and Biometry, F. Proschan and R.J. Serfling, eds., SIAM, Philadelphia, pp 379- 410.
Crow, L.H. (1975), "On Tracking Reliability Growth," Proceedings Annual Reliability
and Maintainability Symposium, pp. 438-443.

Crow, L.H. (1982), "Confidence Interval Procedures for the Weibull Process With
Applications to Reliability Growth," Technometrics, 24(1):67-72.
Crow, L.H. (1990), "Evaluating the Reliability of Repairable Systems," Proceedings
Annual Reliability and Maintainability Symposium, pp. 275-279.
Crow, L.H. (1993), "Confidence Intervals on the Reliability of Repairable Systems,"
Proceedings Annual Reliability and Maintainability Symposium, pp. 126-134
Duane, J.T. (1964), "Learning Curve Approach to Reliability Monitoring," IEEE
Transactions On Aerospace, 2, pp. 563-566.
Gumbel, E. J. (1954), Statistical Theory of Extreme Values and Some Practical
Applications, National Bureau of Standards Applied Mathematics Series 33, U.S.
Government Printing Office, Washington, D.C.
Hahn, G.J., and Shapiro, S.S. (1967), Statistical Models in Engineering, John Wiley &
Sons, Inc., New York
Hoyland, A., and Rausand, M. (1994), System Reliability Theory, John Wiley & Sons,
Inc., New York
Johnson, N.L., Kotz, S. and Balakrishnan, N. (1994), Continuous Univariate
Distributions Volume 1, 2nd edition, John Wiley & Sons, Inc., New York
Johnson, N.L., Kotz, S. and Balakrishnan, N. (1995), Continuous Univariate
Distributions Volume 2, 2nd edition, John Wiley & Sons, Inc., New York
Kaplan, E.L., and Meier, P. (1958), "Nonparametric Estimation From Incomplete
Observations," Journal of the American Statistical Association, 53: 457-481.
Kalbfleisch, J.D., and Prentice, R.L. (1980), The Statistical Analysis of Failure Data,
John Wiley & Sons, Inc., New York
Kielpinski, T.J., and Nelson, W.(1975), "Optimum Accelerated Life-Tests for the
Normal and Lognormal Life Distributins," IEEE Transactions on Reliability, Vol. R-24,
5, pp. 310-320
Klinger, D.J., Nakada, Y., and Menendez, M.A. (1990), AT&T Reliability Manual, Van
Nostrand Reihhold, Inc, New York
Kolmogorov, A.N. (1941), "On A Logarithmic Normal Distribution Law Of The
Dimensions Of Particles Under Pulverization," Dokl. Akad Nauk, USSR 31, 2, pp
99-101.
Kovalenko, I.N., Kuznetsov, N.Y., and Pegg, P.A. (1997), Mathematical Theory of
Reliability of Time Dependent Systems with Practical Applications, John Wiley & Sons,
Inc., New York
Landzberg, A.H., and Norris, K.C. (1969), "Reliability of Controlled Collapse

Interconnections." IBM Journal Of Research and Development, Vol. 13, 3.

Lawless, J.F. (1982), Statistical Models and Methods For Lifetime Data, John Wiley &
Sons, Inc., New York
Leon, R. (1997-1999), JMP Statistical Tutorials on the Web at
http://www.nist.gov/cgi-bin/exit_nist.cgi?url=http://web.utk.edu/~leon/jmp/
Mann, N.R., Schafer, R.E. and Singpurwalla, N.D. (1974), Methods For Statistical
Analysis Of Reliability & Life Data, John Wiley & Sons, Inc., New York
Martz, H.F., and Waller, R.A. (1982), Bayesian Reliability Analysis, Krieger Publishing
Company, Malabar, Florida
Meeker, W.Q., and Escobar, L.A. (1998), Statistical Methods for Reliability Data, John
Wiley & Sons, Inc., New York
Meeker, W.Q., and Hahn, G.J. (1985), "How to Plan an Acceleration Life Test - Some
Practical Guidelines," ASC Basic Reference In Quality Control: Statistical Techniques -
Vol. 10, ASQC , Milwaukee, Wisconsin
Meeker, W.Q., and Nelson, W. (1975), "Optimum Accelerated Life-Tests for the
Weibull and Extreme Value Distributions," IEEE Transactions on Reliability, Vol. R-24,
5, pp. 321-322
Michael, J.R., and Suchucany, W.R. (1986), "Analysis of Data From Censored
Samples," Goodness of Fit Techniques, ed. by D'Agostino, R.B., and Stephens, M.A.,
Marcel Dekker, New York.
MIL-HDBK-189 (1981), Reliability Growth management, U.S. Government Printing
Office
MIL-HDBK-217F ((1986), Reliability Prediction of Electronic Equipment, U.S.
Government Printing Office
MIL-STD-1635 (EC) (1978), Reliability Growth Testing, U.S. Government Printing
Office
Nelson, W., (1990), Accelerated Testing, John Wiley & Sons, Inc., New York
Nelson, W., (1982), Applied Life Data Analysis, John Wiley & Sons, Inc., New York
O'Connor, P.D.T. (1991), Practical Reliability Engineering (Third Edition), John Wiley
& Sons, Inc., New York
Peck, D., and Trapp, O.D. (1980), Accelerated Testing Handbook, Technology
Associates and Bell Telephone Laboratories, Portola, Calif.
Pore, M., and Tobias, P. (1998), "How Exact are 'Exact' Exponential System MTBF
Confidence Bounds?", to appear in the 1998 Proceedings of the Section on Physical and
Engineering Sciences of the American Statistical Association.

SEMI E10-96, (1996), Standard For Definition and Measurement of Equipment

Reliability, Availability and Maintainability (RAM), Semiconductor Equipment and
Materials International, Mountainview, CA
Tobias, P. A., and Trindade, D. C. (1995), Applied Reliability, 2nd edition, Chapman and
Hall, London, New York

Reliability Notes

Uploaded by

Copyright:

Available Formats

Reliability Notes

Uploaded by

Document Information

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 Notes

Uploaded by

Copyright:

Available Formats

What are the basic terms and models used for reliability evaluation?

What are the basic terms and models used for reliability evaluation?

What are some common difficulties frequently found with reliability data and how are they overcome?

What are some common difficulties frequently found with reliability data and how are they overcome?

8.

Assessing Product Reliability

8. Assessing Product Reliability

3. Reliability Data Collection 4. Reliability Data Analysis

Click here for a detailed table of contents

http://www.itl.nist.gov/div898/handbook/apr/apr.htm (2 of 2) [5/22/2001 1:27:36 PM]

8. Assessing Product Reliability - Detailed

http://www.itl.nist.gov/div898/handbook/apr/apr_d.htm (1 of 4) [5/22/2001 1:27:25 PM]

http://www.itl.nist.gov/div898/handbook/apr/apr_d.htm (2 of 4) [5/22/2001 1:27:25 PM]

5. Empirical model fitting - distribution free (Kaplan-Meier)

3. Reliability Data Collection [8.3.]

4. Reliability Data Analysis [8.4.]

http://www.itl.nist.gov/div898/handbook/apr/apr_d.htm (3 of 4) [5/22/2001 1:27:25 PM]

4. How do you compare reliability between two or more populations? [8.4.4.]

http://www.itl.nist.gov/div898/handbook/apr/apr_d.htm (4 of 4) [5/22/2001 1:27:25 PM]

8. Assessing Product Reliability

http://www.itl.nist.gov/div898/handbook/apr/section1/apr1.htm (1 of 2) [5/22/2001 1:27:37 PM]

http://www.itl.nist.gov/div898/handbook/apr/section1/apr1.htm (2 of 2) [5/22/2001 1:27:37 PM]

8. Assessing Product Reliability

8.1.1. Why is the assessment and control of

http://www.itl.nist.gov/div898/handbook/apr/section1/apr11.htm [5/22/2001 1:27:38 PM]

8. Assessing Product Reliability

8.1.1.1. Quality versus reliability

Life The quality level can be described by a single fraction defective. To

http://www.itl.nist.gov/div898/handbook/apr/section1/apr111.htm [5/22/2001 1:27:38 PM]

8. Assessing Product Reliability

8.1.1.2. Competitive driving factors

http://www.itl.nist.gov/div898/handbook/apr/section1/apr112.htm [5/22/2001 1:27:39 PM]

8. Assessing Product Reliability

8.1.1.3. Safety and health considerations

http://www.itl.nist.gov/div898/handbook/apr/section1/apr113.htm [5/22/2001 1:27:39 PM]

8. Assessing Product Reliability

8.1.2. What are the basic terms and models

http://www.itl.nist.gov/div898/handbook/apr/section1/apr12.htm [5/22/2001 1:27:40 PM]

8. Assessing Product Reliability

8.1.2.1. Repairable systems and non-repairable

http://www.itl.nist.gov/div898/handbook/apr/section1/apr121.htm (1 of 3) [5/22/2001 1:27:41 PM]

1. F(t) = the area under the PDF f(t) to the left of t.

http://www.itl.nist.gov/div898/handbook/apr/section1/apr121.htm (2 of 3) [5/22/2001 1:27:41 PM]

Dataplot Solution: The Dataplot commands for the Weibull are:

http://www.itl.nist.gov/div898/handbook/apr/section1/apr121.htm (3 of 3) [5/22/2001 1:27:41 PM]

8. Assessing Product Reliability

8.1.2.2. Reliability or survival function

http://www.itl.nist.gov/div898/handbook/apr/section1/apr122.htm [5/22/2001 1:27:42 PM]

8. Assessing Product Reliability

8.1.2.3. Failure (or hazard) rate

If we let be the Cumulative Hazard Function then we

http://www.itl.nist.gov/div898/handbook/apr/section1/apr123.htm (1 of 2) [5/22/2001 1:27:43 PM]

It is also sometimes useful to define a single number average failure rate

http://www.itl.nist.gov/div898/handbook/apr/section1/apr123.htm (2 of 2) [5/22/2001 1:27:43 PM]

8. Assessing Product Reliability

8.1.2.4. "Bathtub" curve

http://www.itl.nist.gov/div898/handbook/apr/section1/apr124.htm (1 of 2) [5/22/2001 1:27:44 PM]

http://www.itl.nist.gov/div898/handbook/apr/section1/apr124.htm (2 of 2) [5/22/2001 1:27:44 PM]

8. Assessing Product Reliability

8.1.2.5. Repair rate or ROCOF