Page 1 availability. For example, in the two- time (on the “system clock”) than it would
♦ Critical Examination of a Common component system described above, have if the system had not been down for
Assumption in System Availability suppose that the point availabilities at time the maintenance of another component.
Calculations t for Component 1 and Component 2 are This effect of system operation is not taken
♦ Updating the Classic Reliability Block 80% and 90%, respectively. If the system into account in the estimation of the
Diagram Methodology and Constructs availability is simply the product of the availability for the individual component
component availabilities then the system and yet it is quite relevant to the availability
Page 2 availability at time t would be AS = A1 • A2 = of the system.
♦ From the Editor’s Desk (.80 • .90) = .72 or 72%.
This method can be justified from a Simple Example to Demonstrate the
Page 7 Effect of System Operation on
probabilistic perspective because both
♦ Guest Commentary: How to Specify Component Availability
items need to be available when called
for Reliability The following deterministic scenario
upon in order for the system to be
Page 8 available. However, the method does not demonstrates the effect of system
operation on component availability.
♦ Failure Modes, Effects and Criticality take into account the effect on component
Analysis availability, if any, when the components Consider a system with two units
are operating together in a system configured reliability-wise in series, where
Page 16 configuration. If the component continues Unit 1 fails every 100 hours and takes 20
♦ Guest Submission: Integrating Weibull to operate even when the system is down hours to restore and Unit 2 fails every 75
Analysis into Bodyshop Reliability
due to the failure of another component, hours and takes 25 hours to restore.
Engineering Furthermore, neither component
then the availability will be the same when
♦ Case Study Report: Failure Reporting, continues to accumulate age when the
calculated individually as it is when
Evaluation and Display (FRED) system is down. The individual
Report calculated with respect to its behavior
within the system. However, if the availabilities of the components for 300
Page 20 component does not continue to operate hours are 86.6% and 75% respectively, as
♦ Resources for the Reliability when the system is down, then its shown in Figures 1 and 2 (page 3).
Professional on weibull.com availability will be different within the However, when we analyze the
system than when calculated individually. components operating together in a
Page 23 This is because when the system is down system, we see that Unit 2 will fail first at
♦ For Your Information 75 hours, causing the system to fail. The
due to the failure of another component,
♦ Bulletin Board the given component will not accumulate system will then be undergoing
age. Therefore, although the “system maintenance for 25 hours and will be
clock” will advance by the amount of time operational again at 100 hours. At 125
Reliability Edge is published four times a year. To that it takes to restore the other hours, the system will fail again, this time
obtain a free subscription, to send comments or to
component, the “component clock” for the due to Unit 1. This is because Unit 1 fails
submit articles for consideration:
given component will not advance and the after 100 hours of operation and it had
ReliaSoft Publishing
ReliaSoft Plaza component will be likely to fail at a later Please Turn to Page 3
115 S. Sherwood Village Drive
Tucson, AZ 85710 USA
Telephone: +1.520.886.0366 Fax: +1.520.886.0399
E-mail: ReliabilityEdge@ReliaSoft.com
From the Editor’s Desk...
This issue of Reliability Edge includes To submit, please prepare the article
Correspondence with the editor may be published, in two guest submission articles: a case in Microsoft Word and e-mail it to
whole or in part, in future issues of ReliaSoft
publications. study from Dave Whetton to demonstrate ReliabilityEdge@ReliaSoft.com. We will
An electronic copy of this document can be viewed an effort to integrate Weibull analysis into review the article and will notify you of our
or downloaded from: www.ReliaSoft.com/newsletter bodyshop engineering (page 16) and a intention to publish, then cooperate with
For information about products and services:
guest commentary on reliability you on any proofreading/editorial changes
ReliaSoft Sales specifications from Thimmiah Gurunatha that may be required. If the article includes
ReliaSoft Plaza (page 7). We appreciate their willingness graphics, we may request that you provide
115 S. Sherwood Village Drive
Tucson, AZ 85710 USA
to share their experiences with colleagues the original graphics in electronic format.
Telephone: +1.520.886.0410 Fax: +1.520.886.0399 and invite you to do the same. Prior to publication, you will be asked to
E-mail: Sales@ReliaSoft.com We welcome guest submission review and approve the final version of the
Web Site: www.ReliaSoft.com
articles on topics of interest within reliability article as it will appear in
This document may be reproduced without engineering and related fields. In general, Reliability Edge and to
permission provided that it is not altered in any way
and all pages are included in any reproduction.
we expect guest submission articles to be sign and submit a
relevant, readable and theoretically sound. publication release form.
©2002 ReliaSoft Corporation, ALL RIGHTS RESERVED.
ReliaSoft, Weibull++, Weibull.com, Reliability Edge, ALTA, The decision to publish (or not publish) a
BlockSim and RG are registered trademarks of
ReliaSoft Corporation.
specific article is made by our editorial
review panel. --Lisa Hacker
Quarter 3, 2002 Page 3
Continued from Page 2: “Critical Examination of a Common Assumption in System Availability Computations”
accumulated 75 hours before system failure and another 25 Mathematical Demonstration
after the system was restored. The same process can be The following example mathematically demonstrates the
repeated yielding the system results shown in Figure 3. effect of the assumption that the system availability is always
As the example demonstrates, the distinction of whether a simply the product of the component availabilities.
component continues to operate when the system has failed Consider a very simple system with two identical
or if a component can fail when the system is undergoing repair components configured reliability-wise in series, and with
is very important and needs to be taken into account when constant failure and repair rates. Both components cease to
performing such analyses. Likewise, the computation used to operate when the system is down. (Note: We assume the
estimate system availability must take into account the exponential distribution for both times-to-failure and times-to-
availabilities of the components with respect to their operation repair in order to simplify the calculations, while recognizing
within the system. The component availabilities determined that the exponential distribution is rarely appropriate to describe
individually and with respect to system operation will be the failure behavior in the real world. The term “repair” is used to
same only in situations where the component is not affected describe the maintenance action required to restore the
by the failure of other components and/or the system (or where component to operation and we assume that the component is
no other components fail within the given time). “good as new” as a result of the maintenance action.)
Eqn (1)
Figure 2: Up and Downtimes for Unit 2 Individually Since we are using a simple exponential distribution for
both the failure and repair distributions, it is easy to determine
the system availability using another methodology such as
Markov analysis. By comparison with the Markov result, we
can evaluate the assumption that the system availability is the
product of the component availabilities. The first step in the
Markov approach is to determine the possible system states.
These are:
Authors
Unless otherwise attributed, the articles in Reliability Edge
have been developed by ReliaSoft’s R&D staff. This dedicated
team of engineers, statisticians, mathematicians and
programmers works continuously with top experts in the
discipline to develop principles and theory that significantly
advance the current state of research and with industry
practitioners to successfully apply those principles in the field.
Contributing authors hold advanced degrees in Reliability
Figure 4: Markov Diagram for the Simple System Engineering and related fields.
Quarter 3, 2002 Page 5
Continued from Page 1: “Updating the Classic Reliability Block Diagram Methodology and Constructs”
function. The concept can be expanded further, as shown in (such as the diagram in Figure 3) and thus require a more
Figure 3, with combinations of series and parallel configurations advanced analytical treatment. A network is a good example
in the same diagram. These elementary configurations form of system requiring a “Complex” reliability block diagram and
the basis of the reliability block diagram constructs. Figure 4 illustrates an example of this RBD type.
Complex Configurations K-out-of-N Nodes
If one takes the approach a step further, “Complex” block The creation of reliability block diagrams with series,
diagrams can be created. “Complex” diagrams cannot be parallel, combination and complex configurations required the
expressed as a simple combination of series and parallel blocks use of blocks and lines only. To extend the functionality of a
block diagram, one needs to
introduce some additional elements
to the “tool kit.” One such element is
the k-out-of-n node, which allows the
analyst to specify an alternative form
of redundancy known as k-out-of-n
redundancy. A k-out-of-n node can
have n paths leading into it, and
requires that k of those n paths must
function for the system to function.
Figures 5 and 6 present RBDs that
incorporate such nodes. The
configuration in Figure 5 includes a
Figure 3: Combination of Series and Parallel Configurations Please Turn to Page 6
FigureFigure 4: Complex
6: Complex System Configuration
Configuration with 2-out-of-5 Node
Page 6 Quarter 3, 2002
Continued from Page 5: “Updating the Classic Reliability Block Diagram Methodology and Constructs”
k-out-of-n node where either B, C or D must operate for the when needed. Figure 7 includes a standby container with three
system to function but any two of the other items may fail without items in standby configuration where one component is active
causing system failure. With the traditional reliability block while the other two components are idle. One block within the
diagram methodology, k-out-of-n redundancy could have been container must be operating or, because the container block is
specified for units drawn in parallel. However, with the part of a series configuration, the system will fail.
introduction of the node element in the diagram, one can specify
such redundancy for complex configurations, such as the one Load Share Containers
shown in Figure 6 (page 5). The container concept can be expanded to also represent
load sharing configurations. As the name implies, load sharing
Standby Containers configurations consist of a components that are in load sharing
A Standby Container can be used to represent items redundancy. Units in load sharing redundancy exhibit different
configured with standby redundancy. Standby redundancy failure characteristics when one or more fail. In Figure 8, units
configurations consist of items that are inactive and available 1, 2 and 3 are in a load share container and have their own
to be called into service when/if the active item fails (i.e. on failure characteristics. All three must fail for the container to
standby). A container block, with other blocks inside, is utilized fail. However, as individual items fail, the failure characteristics
to better achieve and streamline the representation and analysis of the remaining units change since they now have to carry a
of standby configurations. The container serves a dual purpose. higher load to compensate for the failed ones.
The first purpose is to clearly delineate and define the standby In addition to Standby and Load Share containers, other
relationships between the active unit(s) and standby unit(s). new types of blocks can be used to increase the versatility of
The second purpose is to serve as the manager of the switching reliability block diagram constructions and facilitate more rapid
process. For this purpose, the container can be defined with creation of diagrams that are easier to read. These include
its own probabilities of successfully activating standby units Subdiagram Blocks, Multi Blocks and Mirror Blocks, which are
described next.
Subdiagram Blocks to Represent Inheritance
A Subdiagram Block inherits some or all of its
properties from another block diagram. This allows the
analyst to maintain separate diagrams for portions of a
system and incorporate those diagrams as components
of another diagram. With this technique, it is possible to
generate and analyze extremely complex diagrams
representing the behavior of many subsystems,
subsubsystems etc. in a manageable way. In Figure 9,
Subdiagram Block A in the top diagram represents the
series configuration of the subsystem reflected in the
middle diagram, while Subdiagram Block G in the middle
diagram represents the series configuration of the
subsubsystem in the bottom diagram.
Figure 7: Configuration with Standby Container Please Turn to Page 7
Figure 8: Configuration with Load Share Container Figure 9: Subdiagram Blocks to Represent Inheritance
Quarter 3, 2002 Page 7
Continued from Page 6: “Updating the Classic Reliability Block
Diagram Methodology and Constructs” Guest Commentary
Multi Blocks to Save Time and Space How to Specify for Reliability
By using Multi Blocks, a single block can represent multiple
identical blocks in series or in parallel configuration. This Thimmiah Gurunatha
technique is simply a way to save time when creating the RBD Principal Engineer - Xerox Corporation
and to save space within the diagram. Each item represented TG Rainbow Incorporated
by a Multi Block is a separate entity with identical reliability PE, CRE, CSSB and Six Sigma Master Black Belt - ASQ
characteristics to the others. However, each item is not rendered
individually within the diagram. In other words, if the RBD As an American Society of Quality Six Sigma instructor
contains a Multi Block that represents three identical and mentor, I have observed abundant evidence that the
components in a series configuration, then each of those effective utilization of reliability specifications can be a
components fails according to the same failure distribution but challenge to engineers. To address this challenge, I propose
each component may fail at different times. Because the items for your consideration a simple step-by-step approach to
are arranged reliability-wise in series, if one of those implement and integrate reliability specifications:
components fails, then the Multi Block fails. It is also possible
to define a Multi Block with multiple identical components • Benchmark current reliability numbers for the function you
arranged reliability-wise in parallel or k-out-of-n redundancy. are replacing in your new design.
• Make sure that you consider both your past products and
the products of your competitors.
• Understand the current failure modes of the best product
and innovate to improve the reliability.
• Ensure that you do not create new failure modes.
• Based on the leading failure mechanisms of the system
Mirror Blocks to Simulate Bi-Directional Paths or component, determine the survival distribution by
While Multi Blocks allow the analyst to represent multiple testing. I recommend accelerated testing.
items with a single block in the RBD, Mirror Blocks can be • Specify the 2%, 50%, 63.2% (Characteristic Life) and 98%
used to represent a single item with more than one block placed
failure points in hours, days, cycles or years, as
in multiple locations within the diagram. Mirror Blocks can be
applicable.
used to simulate bi-directional paths within a diagram. For
example, in a reliability block diagram for a communications • Ensure that you call out the real stresses under which
system where the lines can operate in two directions, the use the component or subsystem is used. This is critical for
of Mirror Blocks will facilitate realistic simulations for the system getting good correlation of test data to real life.
maintainability and availability. It may also be appropriate to • Determine critical parameters that drive the quality and
use this type of block if the component performs more than reliability.
one function and the failure to perform each function has a • Specify the reliability and critical specification of
different reliability-wise impact on the system. parameters that drive the failure mechanisms. Parameter
specification comes from characterization and
optimization test results.
• Include a warranty note in the drawing to enable proper
controls at the supplier for the quality today, tomorrow
and thereafter.
Conclusion • Ensure that proper field returns procedures are in place
As this article demonstrates, it is possible to update and to find root cause of failures.
expand the techniques available for reliability block diagrams • Ensure that you have resources and tools available for
in order to increase the effectiveness of the analysis tool and immediate feedback to the source of problems.
also improve the ease of creation and appearance of the
To demonstrate the effectiveness of this methodology, I
diagrams. The proposed enhancements to the methodology
would like to conclude with an example of its implementation.
include complex configurations, k-out-of-n nodes, Standby and
The example involves a clean sheet design of a complex
Load Sharing containers, Subdiagram Blocks to represent
product intended to replace the existing fleet with lower cost
inheritance from other diagrams, Multi Blocks to represent
units at no extra service cost. When the activity started, the
multiple identical blocks and Mirror Blocks to represent the
actual was 1 million cycles and the target was 5 million cycles.
same block in multiple locations within the diagram. ReliaSoft’s
By implementing the methodology described above, the
BlockSim 6 software supports all of the standard RBD
company was able to achieve a design with predicted reliability
techniques as well as the enhancements discussed in this
from accelerated testing of 26 million cycles. For one
article. On the Web at http://BlockSim.ReliaSoft.com.
component, we were able to improve the reliability by 26 times
RS at no extra cost to the component, two years before launch!
Page 8 Quarter 3, 2002
For example, consider the following partial FMEA for a • Probability of Loss (PL): The probability that the failure
battery, which uses ten point rating scales to rank the severity, mode will cause a system failure (or will cause a significant
occurrence and detection: loss). This is an indication of the severity of the failure effect
and may be set according to the following scale:
Figure 2: Xfmea interface with item and FMEA hierarchies Figure 3: Xfmea Failure Cause Properties window
A NALYZE ACCELERATED LIFE DATA
WITH UP TO 8 SIMULTANEOUS STRESS
TYPES OR WITH STRESS
PROFILES THAT V ARY
WITH TIME
Life-Stress Relationships
-Arrhenius
-Eyring Quantitative Accelerated Life Testing
-Inverse Power Law In a quantitative accelerated life test, the stress(es) that cause product failure are accelerated under
-Temperature-Humidity controlled conditions to cause the product to fail in a shorter period of time. This testing approach
-Temperature-NonThermal
produces time-to-failure information for products more quickly but does not introduce additional failure
-Proportional Hazards
mechanisms that would not occur under normal use conditions. Mathematical models can be used
-General Log-Linear
-Cumulative Damage to extrapolate a use-level cumulative distribution function (cdf) for the product from the life data
Time-Dependent Stress Profiles obtained under accelerated conditions. Quantitative accelerated life data analysis provides reliability
Life Distributions and probability of failure, warranty time, mean life, B(X) life and other reliability metrics for the
-Weibull product under normal use conditions from accelerated life data.
-Exponential
-Lognormal
Maximum Likelihood Estimation ALTA 6 and ALTA 6 PRO
Confidence Bounds The mathematical models required to analyze data obtained from
Calculated Results quantitative accelerated life tests are as complex as they are powerful.
-Reliability & Probability of Failure ReliaSoft's ALTA is the first and only commercially available software
-Mean Life package designed expressly for quantitative accelerated life testing
-Warranty Time Information data, using rigorous scientific analysis methods.
-B(X) Life
-Acceleration Factor
This product, the next generation of the ALTA
-Failure Rate
Automated 2D and 3D Plots software, is available in two versions: Standard
-Probability (ALTA 6) and Professional (ALTA 6 PRO). Both
-Reliability vs. Time versions provide an enhanced interface, expanded
-Unreliability vs. Time plotting and reporting capabilities, integration with
-Probability Density Function (pdf) Weibull++ 6 and additional utilities for degradation
-Failure Rate vs. Time analysis and other supplementary analyses. In
-Life vs. Stress addition, ALTA 6 PRO provides new life-stress
-Standard Deviation vs. Stress relationship models that allow you to analyze data
-Acceleration Factor vs. Stress
from tests with time-dependent stress profiles
-Residuals Plots
or with up to eight simultaneous stress types.
-3D Likelihood Function
-3D Failure Rate vs. Stress
-3D pdf vs. Stress Minimum System Requirements: Windows 98, Me, NT, 2000 or XP. Pentium class processor with 32 MB
-3D Reliability vs. Stress RAM, SVGA display and at least 35 MB of hard disk space.
-3D Unreliability vs. Stress Documentation: Complete and detailed on-line help, as well as printed product documentation and a reference
Degradation Analysis book that presents accelerated life testing analysis principles and theory.
Integration with Weibull++ 6 Support: ReliaSoft’s unparalleled after-sale support includes free telephone, fax and e-mail support and free
Customized Analyses minor version updates.
Automated Reports and Graphs ALTA 6 Standard Price: $995. Upgrade from ALTA 1.0 for $395.
ALTA 6 PRO Price: $4,995 (includes free training course). Upgrade from ALTA 1.0 or from
ALTA 6 Standard for $4,195 (includes free training course).
+1.520.886.0410 http://ALTA.ReliaSoft.com
Quarter 3, 2002 Page 11
Continued from Page 9: “Failure Modes, Effects and Criticality Analysis”
For example, consider a criticality analysis for the partial for the system/process. For example, the FMEA can be
FMEA on the battery. The reliability of the battery is 92% at the incorporated into an effective Reliability Growth management
operating time of interest, therefore the unreliability is 8%. The policy by providing a structure to organize information about
portion of the item unreliability that can be attributed to the product failures and assisting with efforts to identify the failure
given failure mode is 25% (or 25% of the item failures are likely modes that have been observed during reliability growth testing
to be due to this particular failure mode). The probability of and the failure modes that may still yet be observed.
loss is 100% because the occurrence of the failure mode will
cause a system failure. The Criticality for the failure mode is Conclusion
(.08)(.25)(1.00) = .02, or 2%. As with the RPN method, this FMEA/FMECA analysis is a flexible process that can be
Criticality value can be compared with the Criticalities for other adapted to meet the particular needs of the industry and/or the
failure modes to help rank the issues that must be addressed. organization. However, most analyses include the basic
procedures and data requirements described in this article.
Item Q Function Failure FMFR Effect PL C au se Cr
ReliaSoft’s Xfmea software supports these basic procedures,
Battery .08 Provide Fails to .25 System 1.00 Battery .02
the major published industry standards (e.g. J1739, MIL-STD-
adequate provide fails to plates are
relay adequate operate shorted 1629A, etc.) and also provides the flexibility to customize the
voltage power analysis and reports to meet the user’s needs for a particular
application. On the Web at http://XFMEA.ReliaSoft.com.
Figure 4 displays the Xfmea Criticality Analysis utility, which
can also be used to generate FMECA charts and reports. FMEA/FMECA References
Many references for FMEA/FMECA analysis are available
Applications and Related Analyses in print and on the Web. Some useful references include:
FMEA/FMECA techniques are used throughout industry
for a variety of applications and the flexible analysis method • Kececioglu, Dimitri, Reliability Engineering Handbook
can be performed at various stages in the product life cycle. Volume 2. Prentice-Hall Inc., Englewood Cliffs, New Jersey,
FMEA/FMECA analysis can be employed to support design, 1991. Pages 473-506.
development, manufacturing, service and other activities to • MIL-STD-1629A: Procedures for Performing a Failure
improve reliability and increase efficiency. For example, there Mode Effects and Criticality Analysis. U.S. Department of
is widespread use of both design and process FMEAs within Defense, Washington, D.C., November 28, 1984. Note:
the automotive industry and documentation of this analysis is This standard was cancelled by the DoD in August 1998.
a common requirement for automotive suppliers. This
• SAE Aerospace Recommended Practice ARP5580:
methodology is also widely used in the aerospace, medical
Recommended Failure Modes and Effects Analysis (FMEA)
and other manufacturing industries.
Practices for Non-Automobile Applications. SAE
The MSG-3 procedures used by the airline industry
International, Warrendale, PA, 2001.
incorporate FMEA techniques into the analysis procedure.
(Reliability Edge Volume 3, Issue 1 contains an article on MSG- • SAE Surface Vehicle Recommended Practice J1739:
3 and ReliaSoft’s MPC 3, software designed to automate the Potential Failure Mode and Effects Analysis in Design
process.) Likewise, Reliability Centered Maintenance (RCM) (Design FMEA), Potential Failure Modes and Effects
procedures incorporate FMEA as a primary component of the Analysis in Manufacturing and Assembly Processes
analysis. (Process FMEA), and Potential Failure Mode and Effects
In addition, the FMEA reporting structure can be used to Analysis for Machinery (Machinery FMEA). SAE
provide a centralized location for reliability-related information International, Warrendale, PA, June, 2000.
• Stamatis, D.H., Failure Mode and Effect Analysis: FMEA
from Theory to Execution. American Society for Quality
(ASQ), Milwaukee, Wisconsin, 1995.
RS
Reliability Training
ReliaSoft's training seminars provide instruction in reliability
engineering principles and theory as well as the software tools
designed to put that theory into practice. Upcoming seminar
dates include November 18 - 22, 2002, February 24 - 28, 2003,
May 19 - 23, 2003, November 17 - 21, 2003 and February 23 -
27, 2004 in Tucson, Arizona. On-site seminars are also available
as well courses on other reliability topics. Please consult the
Web or contact ReliaSoft for more information.
Figure 4: Xfmea Criticality Analysis utility On the Web at: http://Seminars.ReliaSoft.com
Over the past three years, ReliaSoft's scientists and engineers, along with
corporate partners and selected user groups, have quietly worked to create
the ultimate system visualization and analysis tool. The result of
this effort is a monumental breakthrough encompassing
astounding analytical power while maintaining an
artful simplicity of use: BlockSim 6.
Also exclusively in BlockSim, you can enter cost and feasible reliability growth information to
determine the most cost-effective component reliability allocation strategy to meet a
system reliability goal.
The software has been designed to allow multiple users to work cooperatively on the analysis and
provides simple techniques to share portions of an analysis with other users as well as revision tracking
mechanisms. In a networked environment, you can assign permissions to allow more than one authorized
user to access the same database.
System Requirements: Windows 98, Me, NT, 2000 or XP. Microsoft Office 97 or higher.
Support: ReliaSoft’s unparalleled after-sale support includes free telephone, fax and e-mail support.
Free minor version updates are also included.
Case Study Report: Failure Reporting, Evaluation and Display (FRED) Report
Station 003 Front Structure software applications were used to produce the FRED charts,
Jaguar Castle Bromwich as prescribed by Point 1 of the Ford Corporate R&M
specification.
PROJECT – X202
Bodyshop Methodology
30/8/02 The raw data set used was from the equipment fault detail
report of POSMON (the facility’s event tracking system)
beginning 07:30 on 28th May 2002 and ending 12:00 on 19th
Objective(s) July 2002. The study included all components in Station 003,
The primary objective of this FRED report is to identify an as shown in the BlockSim block diagram in Figure 2 (page 17).
opportunity to improve and raise the baseline of acceptable During this period, 882 events were recorded in the specified
reliability levels of the production equipment. This reliability shift pattern. Events were analyzed at three levels: Station,
growth can only be accomplished through root cause analysis Component (e.g. Rivet Gun, Proximity) and Fault Code.
and then by engineering a sound, verifiable fix. To make the POSMON downtime log readable by
Weibull++ MT, some pre-processing was required. First, the
Introduction data set was filtered to remove any events that fell outside the
The following report represents an Availability study of the specified shift pattern and then events were removed that had
X202 Front Structure Station 003 in the Jaguar Castle Bromwich a continuous duration over planned downtime. The data set
Plant. ReliaSoft’s Weibull++ MT and BlockSim MT (version 1) Please Turn to Page 17
Quarter 3, 2002 Page 17
Continued from Page 16: “Case Study Report: Failure Reporting Evaluation and Display (FRED) Report”
was then entered into the Weibull++ data entry form. For the Conclusions
Weibull analysis, the Rank Regression estimation method was The immediate opportunity for reliability improvement
used because of the completeness of the data. Time-to-failure follows a critical path of Component = Proximity (98.53%
and time-to-repair (or failure duration) distributions with their availability) to Fault Code = 102062 waiting part present (98.5%
parameters were then used to build a reliability block diagram availability).
in BlockSim. Within BlockSim, a simulation was run for 10,000 The study has highlighted the importance of having data in
minutes (approximately 2 weeks of production) with results electronic format that can be operated on for pre-processing
calculated for instantaneous availability 10 times. The simulation before entry into the analysis software. The study also
resolution was set to run 100 inner loops and 10 outer loops. demonstrates the value of the ReliaSoft software for performing
This means that 100 simulation points were generated for each the analysis and creating graphical representations of the
reliability entity and results were returned for each of the 100 output. Finally, the study illustrates the value of automated
runs; then the simulation returned results 10 times at system analysis to point out the dominant areas and sources of failure
level, each time with a new stream of random numbers for the as an aid to prioritize reliability improvement efforts.
simulation points. This yielded 10 different system reliability For Comau Estil UK’s throughput simulation model, the
values and 100 reliability entity values. The system reliability most easily understood input would be the MTBF from the
at the specified time returned by the simulation was the average FRED tree generated by BlockSim. It would be preferable to
of these 10 reliability values. The simulation results were used use MCBF, which represents the failure frequency based on
to estimate MTBF, MTTR and Availability at the System level. station/component busy time.
RS
Similar steps were taken to work down the availability
hierarchy from Component level to Fault Code level, as shown
on the FRED chart in Figure 1.
Results
Figures 1 through 4 demonstrate the results of the analysis.
Figure 2: Front Structure Station 003 Block Diagram Figure 4: Front Structure Station 003 Rivet Gun FRED Tree
RAC is ready to provide total, turn-key solutions including
training, data, consulting services, information and support
for both government and commercial businesses.
consultation and reliability data
experience publications
implementation and software
The Reliability Analysis Center RAC’s technical staff is ready to As the central resource for DoD and
(RAC) is a Department of Defense provide support at all levels and in all industry reliability data, RAC offers
(DoD) Information Analysis Center areas of reliability. Organizations the data that’s important to your
operated by IITRI for more than which already have relatively mature business, and we integrate this
33 years. The DoD has chartered reliability programs have RAC work information into our products,
RAC as a Center of Excellence and on very specific problems or issues. including our publications and
a scientific and technical resource for Others have no formal reliability software. Our extensive line of
information, data, analysis, analytical program and RAC provides a wider reliability publications offers
tools, training and technical range of support beginning with an something for everyone: managers
assistance in the broadly defined assessment of the organization’s achieve a high-level understanding of
fields of Reliability, Maintainability, reliability needs. Depending on a the relationship between reliability and
Supportability and Quality. customer’s needs, RAC will provide the bottom line, design engineers
Although chartered by the DoD, short-term, quick responses to receive support for design trade-off
RAC undertakes a variety of other immediate needs, or long-term analyses and decisions, and reliability
government organization and studies and independent analysis. practitioners will find practical
industrial support projects each year. Our philosophy is to bring the application and how-to guides. RAC’s
Our 60+ local engineers are customer’s level of knowledge and software provides reliability
backed by IITRI resources at 15 experience to a point where the professionals with the tools needed to
IITRI locations with over 1300 customer can independently and make accurate, informed decisions and
employees of various technical successfully implement the perform required program activities
backgrounds and expertise. reliability program for the long with efficiency. RAC’s PRISM®
term. After that point, RAC can software sets a new standard in
provide assistance to solve specific reliability prediction as a
training problems or deal with unique issues comprehensive resource that includes
not envisioned in the initial reliability next generation component failure rate
Learn from the experts with RAC’s program plan. models, searchable failure rate
world-renowned professional databases, software reliability
reliability training program. Courses predictions and operating and non-
are taught by practicing professionals operating failure rates.
using relevant experience and real-
world examples. To learn at your
own pace, RAC’s distance learning
is structured for all individuals
working in the reliability engineering
realm, including systems and circuit
design engineers, program managers,
quality engineers and members of
related disciplines.
free services
Figure 1: Quick Statistical Reference (QSR) iTool Figure 2: The SPRT utility, available for free download
Quarter 3, 2002 Page 21
Continued from Page 20: “Resources for the Reliability Professional on weibull.com”
Reliability eTextbooks are currently available for: • Other Resources: Other resources provided by
weibull.com include lists of recommended reliability books,
• Life data analysis (Weibull analysis).
reliability- and quality-related Web links, an on-line glossary
• Accelerated life testing data analysis, with an emphasis of reliability terms as well as a complete “blueprint” for
on quantitative accelerated life testing (QALT) analysis. implementing a successful reliability program. Resources
• System reliability, maintainability and availability analysis, are added and improved on an ongoing basis.
with an emphasis on reliability block diagram approaches.
Discussion Forum
• Reliability growth analysis.
The weibull.com site is provided as a service to the
reliability community and, as such, features an interactive
Reliability References, Case Studies and Other Resources
discussion forum designed to facilitate communication within
Other references on weibull.com include:
the field. Discussion topics include a general reliability
• Quick Subject Guides: The site’s Quick Subject Guides discussion, employment opportunities and questions/comments
provide a brief overview of some of the key topics in regarding ReliaSoft’s standard software products. These forums
reliability engineering to provide a quick and easy-to- are open to everyone and no password is required. ReliaSoft
understand foundation in the subject. These references personnel monitor and participate in the discussions.
also provide suggestions for further, more detailed,
research. Conclusion
• Case Studies: The site provides access to case studies The extensive array of reliability resources described in
for subjects such as life data analysis and accelerated life this article attracts thousands of visitors each month to
data analysis. These case studies include step-by-step weibull.com. The weibull.com site (shown in Figure 3) is
instruction for performing the analyses in ReliaSoft’s updated and enhanced on an ongoing basis and ReliaSoft
software packages. makes every effort to be responsive to requests from the
reliability community. To make suggestions for improvement or
• Reliability Publications: The site provides access to on- to request additional references and tools, contact
line versions of the Reliability Edge quarterly newsletter Webmaster@ReliaSoft.com or complete the simple form
as well as the HotWire monthly eMagazine. In addition, located on the Web at http://www.weibull.com/Feedback.
the site includes a comprehensive subject index to the RS
articles in those publications.
• Research Papers: Scholarly research papers that have Another Reliability Publication
been published/presented by ReliaSoft’s R&D personnel
are available for download in PDF format (*.pdf). In addition to the Reliability Edge quarterly newsletter,
ReliaSoft also publishes a monthly eMagazine for reliability
• Seminars and Training Guides: The site provides access professionals, which is distributed via e-mail and available on
to information on training seminars for reliability engineering the Web. The Reliability HotWire eMagazine provides
and related subjects, as well as training guides in PDF information and tips on how to best improve reliability practices
format (*.pdf). The training guides include step-by-step and get the most out of ReliaSoft’s tools for reliability and life
examples for reliability analyses performed with ReliaSoft’s data analysis. This monthly publication brings practical hints
software products. regarding reliability engineering and life data analysis to the
novice and seasoned practitioner alike. In addition, Reliability
HotWire presents tips on how to get the best use out of
ReliaSoft’s reliability analysis software tools. Recent HotWire
articles include:
• Looking at Thermal Cycling in Quantitative Accelerated Life
Testing (QALT)
• Contour Plots and Confidence Bounds on Parameters
• Verifying the Assumption of a Constant Shape Parameter
in Accelerated Life Data
• Comparison of MLE and Rank Regression When the Data
Set Contains Suspensions
• Financial Applications for Weibull Analysis
• plus numerous tips and techniques for effective use of
ReliaSoft’s software.
Subscriptions are free and can be obtained by contacting
HotWire@ReliaSoft.com or by completing the registration form
Figure 3: Home page for weibull.com on the Web at http://www.weibull.com/hotwire.
Weibull++ 6 builds upon the strengths of
previous versions of ReliaSoft’s
Weibull++ software package, the
standard in reliability life data analysis
for thousands of users and hundreds of
Fortune 500 companies worldwide.
Minimum System Requirements: Windows 98, Me, NT, 2000 or XP. Pentium class processor with 32 MB RAM, SVGA
display and at least 35 MB of hard disk space.
Documentation: Complete and detailed on-line help files, as well as printed product documentation (Weibull++ User’s
Guide) and a reference book that presents life data analysis principles and theory (ReliaSoft’s Life Data Analysis Reference).
Support: Support includes free telephone, fax and e-mail support and free minor version updates.
Price: $795 for a single user. Upgrade from Weibull++ 5.0 for just $295.
For a free evaluation copy or additional information:
Phone +1.520.886.0410
Fax +1.520.886.0399
Toll Free 1.888.886.0410 (USA & Canada)
E-mail Sales@ReliaSoft.com
+1.520.886.0410 http://Weibull.ReliaSoft.com
Quarter 3, 2002 Page 23
http://MPC.ReliaSoft.com
ReliaSoft Corporation Prst Std
ReliaSoft Plaza U.S. Postage
PAID
115 S. Sherwood Village Drive Tucson, AZ
Tucson, AZ 85710 Permit #541
Quarter 3, 2002