the Journal of the First Quarter 1998 Reliability Analysis Center

RAC is a DoD Information Analysis Center sponsored by the Defense Technical Information Center and operated by IIT Research Institute

Pioneering Reliability Laboratory Addresses Information Technology

Written by the RAC Staff with the assistance of the Air Force Research Laboratory Information Directorate history office.

The Air Force Research Another of the first reliability operational parameters into Laboratory Information efforts of RADC was the reliability requirements. The Directorate at the Rome (NY) publication of reliability original team created by Research Site is the current prediction models. An RCA Naresky for reliability name for a laboratory which document, TR 1100, became the technology development was a has been the DoD focal point basis of an RADC technical four man group: Edward for electronic reliability since report which was the first Krzysiak (group leader), the 1950s. Established in 1951 military reference on failure Anthony Coppola, Anthony D. as the Rome Air Development rates of electronic components. Pettinato, Jr., and John Fuchs. Center (RADC), becoming the A series of RADC Reliability As the number of reliability Rome Laboratory (RL) in Notebooks provided updates, personnel grew, statistical December 1990, and officially until RADC became the studies in reliability were adopting its present title in preparing activity for MIL- performed by various October 1997, the organization HDBK-217, Reliability individuals and groups under has been the primary DoD Prediction of Electronic the overall direction of David electronic reliability research Equipment. Prediction models F. Barber and his successor and development agency since for electronic parts became the Anthony J. Feduccia, division reliability was recognized as specialty of Lester Gubbins and, chiefs in a directorate headed an engineering discipline. It on his retirement, Seymour by Naresky. has twice received the annual Morris. award of the IEEE Reliability In 1961, the laboratory began to Society, in 1974 and in 1998. The laboratory also produced create facilities for research in references on failure rates for reliability physics. These One of the first publications non-electronic parts, the considering reliability as an prediction and demonstration of Volume 6 Number 1 engineering discipline was maintainability, and the use of Reliability Factors for Ground Bayesian statistics for Inside this Issue Electronic Equipment, reliability demonstration. It ISSRE ‘97...... 7 published by RADC in 1955. was one of the first to develop Calls for Papers...... 9 It’s author was Joseph J. the concept of system Call for Memories...... 9 Naresky (see biography on effectiveness, a figure of merit New from RAC ...... 10 page 5), who created the combining availability, Industry Brief...... 12 laboratory’s reliability reliability, and capability From the Editor ...... 13 program and developed it from measures into a single measure Letter to the Editor...... 16 a personal commitment to a of the overall worth of a Some Observations on multi-faceted effort by about system to its user. Early studies Demonstrating Availability. 17 100 specialists. on operational influences on Mark Your Calendar ...... 20 reliability foreshadowed more Help from RAC...... 22 recent development RAC Order Form...... 23 of a means for translating desired 1-888-RAC-USER (1-888- 722-8737) facilities included a variety of Circuits) qualification Air Force systems, such as the equipment useful for analyzing committee. More recently, airborne command post, and the the cause of failures in Daniel Fayette led a program “HAVE NOTE” program, under microelectronic devices. These for multichip module (MCM) which the laboratory tested a have been used to isolate and reliability and performance variety of Air Force weapons correct problems in operational assessment. Hybrids, MMIC systems for susceptibility to Air Force systems such as the (Microwave Monolithic electromagnetic interference. Minuteman missile. In 1962, Integrated Circuits), solid- The laboratory also created the laboratory sponsored the state Transmit/Receive numerous test sites in which first symposium on reliability modules, and most recently, aircraft were mounted inverted physics, which has continued Microelectromechanical (MEM) on pedestals so that their under IEEE sponsorship as the devices have all benefited from radiation patterns could be International Reliability laboratory programs. quickly and inexpensively Physics Symposium (IRPS). measured, in comparison to in- Noted leaders of the flight tests. From these With knowledge gained from laboratory’s microcircuit facilities, RADC became known an in-house thin film and reliability and quality as the keeper of the “upside monolithic microcircuit assurance activities through down Air Force.” Air Force manufacturing facility created the years include Joseph aircraft ranging in vintage from by Richard Nelson, the center Vaccaro, Joseph Brauer, the F-4 to the F-22, have produced RADC Exhibit 2867, Edward P. O’Connell, Al appeared on the test stands, Quality and Reliability Tamburrino, Regis Hilow, and recent participants have Assurance Procedures for Charles Messenger, and Dr. been the Navy EA-6B Prowler Monolithic Microcircuits. This Robert Thomas, among others. and some automotive document was the direct In 1994, both the statistical communications systems. Other ancestor of MIL-STD-883, Test studies and reliability physics recent developments have been Methods and Procedures for were integrated into the the use of infrared imaging to Microcircuits, the foundation of Electronics Reliability map electromagnetic fields both military and commercial Division under Eugene without the perturbation of microcircuit quality assurance, Blackburn. The new division conventional probing and of MIL-M-38510, the instituted a program for techniques, and the general specification for integrating diagnostics for development of an electro- microcircuits under the multichip modules permitting magnetic performance monitor military qualified parts efficient chip to system (EMPM) which can record the program, both authored by testability. electromagnetic environment center personnel. The inside a system. Leading EMC laboratory later created MIL- and related studies through the I-38535 General Specification years were Samuel Zaccari, for Integrated Circuits Robert McGregor and Carmen (Microcircuits) Manufacturing, Luvera. MIL-H-38534 General Specification for Hybrid The laboratory has often Microcircuits, and MIL-STD- applied its knowledge of 1772 Certification advanced technologies to create Requirements for Hybrid prototype equipment. In 1967, Microcircuit Facility and Lines, One of the first pictures showing it contracted with General which are the basis of the physics of failure phenomena, this 1960s Electric for the development of RADC photo shows the beginnings of current dual-use qualified corrosion of aluminum circuit MIRAGE (Microelectronic manufacturers system. interconnections. Indicator for Radar Ground Equipment) a technology As reliability studies began in The laboratory has always demonstration model of a the 1950s, the laboratory also been heavily involved in the display having one-tenth the began studies into electro- reliability and quality volume of the UPA-35 radar magnetic compatibility (EMC) assurance of new technologies. indicators in the field, and 100 which resulted in, among other John Farrell, for example, times the reliability. RADC things, models used for chaired the VHSIC (Very was designated the manager for analyzing the EMC problems of High Speed Integrated a joint Air Force -FAA

2 RAC Journal, Volume 6, No. 1 procurement of the FYQ-47 also chaired a committee on a thermal design guide, radar data digitizer, which artificial intelligence reliability management was so superior in reliability to applications to maintenance for manuals, a built-in-test design the FYQ-40 it replaced that a a 1984 study on reliability and guide, and many others. One of requirement for back up units in maintainability by the the organization’s most the field was rescinded. More Institute for Defense Analysis. popular products has been the recent activity included the Since then, led by Dale Reliability Toolkit, first development of the Time Stress Richards, the laboratory has produced in 1988 as the RADC Measurement Device (TSMD) to been involved with the Reliability Toolkit, updated in measure the environmental development of “SMART BIT” 1993 as the Rome Laboratory stresses in localized areas of an applying artificial intelligence Reliability Toolkit, and last equipment, and the integration to fault detection in the Joint produced in 1995 as the of the electromagnetic Surveillance and Target Attack Reliability Toolkit: performance monitor (EMPM) Radar System (JSTARS), and Commercial Practices Edition. into the TSMD. The the development of an R&M The latter provides a guide for development of the RH-32RISC Design Expert System commercial products and (Radiation Hardened 32 bit integrating reliability, military systems under the Reduced Instruction Set maintainability and DoD acquisition reform Computer) is also a recent testability analysis into a policies. Preston MacDiarmid laboratory effort. These efforts software tool capable of (now Director of The were, and are, performed by improving itself by learning Reliability Analysis Center) teams cutting across the from its experiences. initiated the series of toolkits separate reliability offices at and Seymour Morris was the the Laboratory. The laboratory ultimately principle architect of their became the custodian of most of evolution. RADC also pioneered the use of the DoD Specifications, computer aided design for Standards, and Handbooks on The laboratory has been a reliability and the application reliability and consultant in reliability of artificial intelligence as a maintainability. Many of engineering to system program diagnostic aid. Anthony these were written by offices since 1957, when Coppola, lead a Reliability laboratory personnel recognized Coppola initiated support to Panel of the Air Force Project as industry leaders in the the acquisition of the 465-L Forecast II in 1985. Leading a specialties involved, including team of reliability specialists Jerome Klion, Eugene Control System. Later systems from various Air Force agencies Fiorentino, et al. These experts support was managed he proposed and evaluated also produced or technically principally by Anthony D. novel ideas for improving Air guided the production of a Pettinato, Jr., and, on his Force effectiveness. Coppola library of technical references, retirement, Bruce W. Dudley. such as a redundancy notebook, Programs supported included the 412-L European Air Warning and Control System, 414-L Over the Horizon Radar, 425-L NORAD Command and Control System, 474-L Ballistic Missile Early Warning System, 440-L Over the Horizon Radar, and many other large systems. Laboratory personnel were significantly involved with JSTARS which received its baptism of fire in the Gulf War. Laboratory reliability engineers also supported virtually all Air Force ground radar programs and a host of Lee Lopez of General Electric and Capt. Edwin Deady of RADC compare the prototype MIRAGE other smaller equipment display, on table, to the current (1967) standard UPA-35, right. acquisitions.

RAC Journal, Volume 6, No. 1 3 Laboratory personnel were also produced MIL-STD-965, Parts produced a new asked to support developments Control Program. Support in maintainability handbook in by the Wright Laboratory (the this area was led by John response to DoD acquisition SPN/GEANs navigation Farrell and provided to the F- reform requirements for system and the Electronically 16 and Advanced Medium references providing technical Agile Radar) and procurements Range Air To Air Missile guidance rather than by the Aeronautical Systems (AMRAAM) programs, among mandating procedures). Division (the center won an Air others. Force Systems Command award The new Information for its performance in achieving In 1961, the center initiated an Directorate, in fiscal year 98, is unprecedented reliability in electronics parts data completing all commitments in the AN/ARC-164 command repository, to collect, analyze the area of reliability sciences radio, an ASD program). and distribute information and transitioning activities Center personnel were name about the reliability of where appliable towards requested for a series of electronic parts. The initial in- research and development in command level “tiger teams” house facility soon became a information technologies. For convened in the early 1970’s to DoD asset called the esample, the experiences and review reliability progress on Reliability Analysis Center knowledge obtained while avionic systems, including the (RAC) which is now one of 13 working in the field of F-111 Mark II avionics suite information analysis centers reliability are now being and the radars for the F-111, F- (IACs) funded by the Defense applied to development of 4 and F-20 aircraft. Air Force Technical Information Center computer aided design (CAD) Divisions using laboratory (DTIC). Technical direction to tools and techniques for micro- reliability support included RAC is still provided by the electro-mechanical (MEM) the Electronic Systems Division laboratory, specifically by devices; design and advanced (ESD), The Aeronautical Richard Hyle, DTIC’s packaging of wafer scale signal Systems Division (ASD), the Contracting Officer’s Technical processors; design and Space and Missile Systems Representative (COTR). development of adaptable re- division (SAMSO) and the configurable computers and The laboratory has also Armaments Division (AD), all modeling and simulation of pioneered the development of since renamed. information systems. After software engineering tools, fiscal year 98, all reliability The FAA was a joint sponsor including metrics and a related tasks will be referred to with the Air Force for the software engineering other government AN/FYQ-40 and AN/FYQ-47 framework, in another organizations. “Rome radar digitizer developments, laboratory directorate Laboratory” has had a long and was so impressed by the concerned with software and heritage in the area of reliability expertise of the not administered by Naresky reliability sciences and laboratory personnel supporting or his successors. Principle accomplished many significant these, that it requested their contributors in these were Al projects to advance the field of support for FAA radar Sukert, Samuel Dinitto and reliability. The efforts of procurements. Among the FAA Andrew Chruscicki . These “Rome Laboratory” personnel systems supported under an software specialists and wil be missed but their legacy interagency agreement were the Naresky’s reliability will go on. ASR-8, ASR-9 and ASR-10 specialists, notably Fiorentino, collaborated on developing airport radars and the ARSR-2, For more information on models for combining hardware ARSR-3 and ARSR-4 long range technical activities at the and software reliability radars. The Information Directorate of the predictions. Bureau of Mines also obtained Air Force Laboratory at the laboratory reliability support Many of the traditional Rome Research Site, contact under an interagency reliability activities, such as John Bart, AFRL/IF, 36 agreement. maintenance of the Electronic Parkway, Rome NY specifications and standards, 13340. Tel: (315) 330-7701. E- Reliability support activities are being transitioned to other mail: [email protected]. The also included advice on government agencies (although Directorate has a web site at electronic part selection and the organization recently URL: http://www.rl.af.mil. control, based on the center-

4 RAC Journal, Volume 6, No. 1 Joseph J. Naresky: Reliability Pioneer

From these pioneering efforts Engineering Administration and a four man reliability from Syracuse University. group he founded at RADC, he developed a program Mr. Naresky was a member of recognized for its significant Sigma Pi Sigma, a physics contributions and the largest honor society, the IEEE concentration of reliability Reliability Society, specialists in the Department Engineering Management of Defense. In 1980, RADC was Society and Electromagnetic awarded the Air Force Compatibility Group, and the Outstanding Unit Citation for AIAA Committee on its reliability contributions. Reliability and Joseph J. Naresky, the Mr. Naresky’s leadership Maintainability. He was architect of reliability earned him Sustained Superior elected an Associate Fellow of engineering research and Performance Awards in 1959 the AIAA. He served as development at the then Rome and 1966, and the Air Force President of the Reliability Air Development Center Decoration for Exceptional Society, a member of the (RADC) in 1955, was a leading Civilian service, the highest management committee for the figure in reliability engineering award that can be given to an Annual Reliability and until his accidental death in Air Force civilian, in 1968. He Maintainability Symposium, July, 1982. also received the IEEE and U. S. representative on the Reliability Society Award in International Electrotechnical Mr. Naresky’s accomplish- 1974 and was elected an IEEE Commission Technical ments in reliability began when Fellow in 1976. Committee 56 (Reliability). he produced the first general He was also President of the reference manual for reliability Mr. Naresky served with the Rome Rotary Club in 1967 and engineers, Reliability Factors U. S. Army Air Corps during 1968. for Ground Electronic World War II. After the war, Equipment. The document was he was employed by the In 1979, he retired from widely used, including a Army’s Watson Laboratories, government service and joined Chinese translation by the Red Bank, NJ, which was the IIT Research Institute, People’s Republic of China. transferred to Rome, NY as the operator of the Reliability Mr. Naresky also served on the Rome Air Development Center Analysis Center. His last Advisory Group for Reliability in 1951. While at RADC, he project before his untimely of Electronic Equipment, whose earned a Bachelor’s degree in death three years later was to report in 1957 launched Physics and Master’s Degrees in complete the Reliability reliability as an engineering both Electrical Engineering and Design Handbook, now MIL- discipline throughout the HDBK-338. Department of Defense.

Reliability Through the Years at the Air Force Rome, NY Facility: A Sampling

1951 Army Watson 1957 Reliability support Production Ground Electronic Laboratories becomes Rome Air provided to “Big-L” System Equipment, prepared Development Center (RADC) programs 1960 Microelectronics testing 1955 Reliability Factors for 1958 RADC Exhibit 2629 facility established Ground Electronic Equipment uses parts count to establish published reliability requirements and 1961 Requirements for an sequential tests for verification electronic parts data repository 1956 Four man reliability established (led to creation of group organized 1959 MIL-R-26474, RAC) Reliability Requirements for

RAC Journal, Volume 6, No. 1 5 1962 RADC sponsored first 1976 Guidelines for 1990 Thermal analysis reliability physics symposium application of warranties program written for use on formulated personal computers 1963 Reliability data collected on 412-L Air Weapons 1977 Liquid crystals used to 1991 Guide to finite element Control System sites in test and analyze failures in analysis published Germany large-scale integrated circuits 1992 Quantitative 1964 Nondestructive 1978 Design guide for built- evaluation made of screening techniques for in-test published environmental stress screening electronic parts developed (ESS) effectiveness 1979 Reliability and 1965 RADC supported Maintainability Management 1993 Rome Laboratory Weapons Systems Effectiveness Manual published Reliability Engineer’s Toolkit Industry Advisory Committee published 1980 Artificial intelligence 1966 First compendium of applications to testability 1994 Techniques for on-wafer storage failure rates for identified reliability testing of electronic parts published microwave monolithic 1981 Bayesian reliability integrated circuits (MMIC) 1966 RADC Specification test procedures developed for developed 2867 established screening repairable equipment requirements for integrated 1995 Reliability Toolkit: circuits (ancestor of MIL-STD - 1982 Parts derating guide Commercial Practices Edition 883) published published

1967 First microelectronic 1983 RADC sponsors Air 1996 Integrated Diagnostics packaging handbook published Force Academy development of for Multichip Modules (MCM) Bayesian tests for one-shot developed 1968 Minuteman integrated systems. circuit failures analyzed 1997 Rome Laboratory 1984 Design considerations becomes the Air Force Research 1969 Tests of plastic for fault tolerant systems Laboratory Information encapsulated integrated identified Directorate at the Rome site circuits begin 1985 “Smart BIT” concepts 1997 CAD for 1970 Reliability support formulated microelectromechanical provided to F-111 Mark II (MEMs) devices started avionics development 1986 Guide to electronic stress screening (ESS) published 1998 Organization receives 1971 “Tiger Team” reviews second IEEE Reliability Society of avionics systems reliability 1987 Finite element analysis award performed by command request applied to surface mounted package 1972 Antenna measurement facility established (start of 1988 RADC Reliability A complete history of the “upside down Air Force”) Engineer’s Toolkit published organization now designated the Air Force Research 1973 F-16 Avionics 1989 MIL-I-38535 General Laboratory Information Reliability Review Team Specification for Integrated Directorate at Rome will be chaired by RADC Circuits (Microcircuits) available in 1998. Contact: Manufacturing, and MIL-H- Thomas W. Thompson, 1975 Nonelectronic 38534 General Specification for Chief History Office, Reliability Notebook Hybrid Microcircuits published AFRL/IFOIHO, 36 Electronic published Parkway, Rome, NY 13440. 1990 RADC becomes Rome Tel: (315) 330-2757. Laboratory

6 RAC Journal, Volume 6, No. 1 ISSRE ’97 By: Ellen Walker, Attendee

ISSRE ’97, the 8th International Automated SRE Company This is not unlike the phrase Symposium on Software Wide “ given by James Tierney “Do the Right Thing Right the Reliability Engineering, was from Microsoft, a co-sponsor of First Time” which has been held Nov 2-5, 1997 in beautiful, the conference. (“Automated” spoken so often in total quality sunny and warm Albuquerque, refers to the automation of training courses given in the New Mexico. It was attended testing and measurement.) last ten years. It is not unique to by more than 150 participants Terney claims that SRE has software engineering. from government, industry, and had great success at Microsoft academia with a strong and that improved customer Presentations showing from the usage data and useful Underlying threads of the telecommunications industry, predictions of ship dates have presentations ranged from “Yes, and from the research been invaluable. He later you can test reliability into community. Approximately stated however that adoption your software” to “You need to one-third of the attendees were of SRE is closer to 50% than design reliability in because by presenters. Other attendees 100% - leaving the the time you get to test it’s were primarily leaders in interpretation of “great success” often too late or too expensive quality or testing departments to be pondered. to test it in.” There was general of their respective consensus that you need a way organizations. Of the 40 or Tuesday’s Keynote, “Software to determine when the software more presentations given in Reliability in Theory and is reliable and that is three days, 80% addressed Practice” was given by Larry accomplished only through research in Software Dalton, Manager, High testing and tracking failures Reliability Engineering (SRE) Integrity Software Systems throughout the testing with only a small number Engineering, Sandia National interval. Although much addressing actual experience. Laboratories, Albuquerque, discussion centered around how NM. His observation is that in to structure the testing, all the The opening keynote address surety critical applications experts seemed to be in was given by Dieter Rombach, such as nuclear weapons control, agreement that fault intensity Director of the Fraunhofer software-based systems are to is the primary metric and Institute for Experimental be avoided, and if unavoidable, should be measured relative to Software Engineering, then “expect the unexpected” the testing interval in terms of Kaiserslautern, Germany and and make provisions to protect time or natural units. While it was titled “Inspections and against it. He described is difficult for many software Testing: Core Competence for dousing an aircraft (loaded engineers to accept this Reliability Engineering”. The with a nuclear weapon) with paradigm as relevant for talk focused on the use of 1000 gallons of jet fuel and software, it is nevertheless systematic inspections for early setting it on fire as a test to asserted to be necessary for defect detection and the use of determine behavior of the determining software testing for reliability detonator in the “unexpected” reliability. assessment and prediction. realm. He ended his keynote Much of what was said is also with Dalton’s Axioms for In most sessions, regardless of proposed in the “cleanroom” Reliability in Theory and the main emphasis of the topic, approach to software Practice: some form of reliability growth development and embodies modeling was used to determine some of the principles of total Specify the RIGHT THING when the software would meet quality management although a pre-determined reliability the “TQM” term was not Construct the THING RIGHT objective. mentioned. The THING may fail, so Another reoccurring thread in His address was followed by a reduce the consequences many of the presentations was keynote titled “Launching the concept of an “operational

RAC Journal, Volume 6, No. 1 7 profile” driving the testing. In for what they perceive to be software before it is an organization adapting their part of the big picture. ready. cleanroom methodology this Standards for interoperability • Software Engineers do not might be called a “statistical are becoming more important. have equal status with usage probability distribution”. hardware or system If the organizational culture Concerns of Constituents engineers and their has roots in TQM, this might be Some participants were recommendations are not referred to as focusing on the expecting to hear about taken seriously, or they customer. In essence it reflects practical experiences in are not consulted at all. how the software is used rather predicting and measuring • It takes too much time to than how it is built. The software reliability in do the detailed profile weights functionality safety/surety critical requirements definition according to frequency of use or applications. None were and testing needed to criticality of the function. presented. A gentleman who ensure the software is Weighting is then used to works for a company that reliable. If we did that determine the quantity and makes pacemakers seemed very we would never make types of test cases that will confused by the reference to the deadlines, and our comprise testing, insuring that operational profile and a products would cost too the most important or most comment that one can often much. serious errors will be detected negotiate reliability objectives • There is little confidence early in the test cycle. with the customer. in the reliability metrics. Why waste time Major Challenges Few, if any, presentations measuring if the results There were very few sessions delineated how a specific have little value to the that addressed software reliability objective was set decision-makers? reliability relative to system and then actually tracked to reliability. The thrust of one attainment. It is not clear As I reviewed in my mind the session was that true whether this was due to the various strategies and research assessment of software reviewing process, or simply to presented I discovered that reliability is very complex and a lack of papers addressing very little of what was we really can’t have much actual implementation of presented is “new”. We really confidence in the results given software reliability do know how to test. We know our current technology usage, so engineering. how to gather requirements, assume the worst and ensure and who to communicate with that the system as a whole can Concluding Thoughts in requirement definition. We tolerate an uncertain software Throughout the conference I know that our organizational reliability element. found myself relating topics culture impacts what we Determining what software under discussion to an actually do versus what we reliability number to plug into underlying question posed by know we should do. Technology the system reliability equation guest speaker Gerry Weinberg, is providing us with new is perhaps still a mystery. renown software development challenges so fast that we are consultant and author of caught up in it and the basic Software integration and its “Software Reliability, Why principles of software impact on reliability was Aren’t We Doing What We engineering often go by the addressed. Software products Know How To Do? “ He asked wayside. are no longer used in isolation. the question “Why don’t we They must work with a build reliable software?” of Is ISSRE ’97 simply an instance multitude of other software conference attendees and got a of IEEE “preaching to the products from competing plethora of excuses that boiled choir” in that it is attended vendors. Who is responsible for down to the following: primarily by the research the reliability of the community and not by software integration? Customers are • The business area is not people with real world often left in a vacuum because aligned with the problems seeking real world no one owns the integration. No technical area. solutions? Is there another one is responsible for the big • External issues force forum for addressing practical picture. Each vendor accounts organizations to release software reliability issues than

8 RAC Journal, Volume 6, No. 1 this conference? Or is it simply Calls for Papers categories of papers: full reflecting the possibility that papers to be published in the proportionately few software 1998 Military/ Aerospace symposium proceedings and engineers are concerned (Transportation) COTS short topics to be discussed at specifically with software Conference, to be held the conference but not reliability? Is software August 26-18, 1998 in published. Abstracts are due by reliability engineering Albuquerque, NM, seeks papers May 4, 1998. Submittal details actually being practiced by the on assuring the highest quality, were not available at receipt of software community at large? availability, reliability and this announcement, but will be Do the “real world” software cost effectiveness in micro- posted on the Shock and people use other names for electronic technology and its Vibration Information Analysis software reliability issues? insertion into high per- Center (SAVIAC) website at formance, affordable systems. URL: In closing, in spite of the concern Authors are requested to submit http://saviac.xservices.com, over attendance, and the fact five copies of a one-page and published in future issues of that the conference was abstract by May 31, 1998 to: the SAVIAC Current research focused, I came back Edward B. Hakim Awareness Newsletter. For with a sense of personal The Center for Commercial subscription to the latter, mail responsibility to my role as a Component Insertion, Inc. request to SAVIAC, 2231 software engineer that I did not 2412 Emerson Ave. Crystal Drive, Suite 711, have before the conference. I Spring Lake NJ 07762. Arlington VA 22202, or Fax to simply have to start “doing Tel. and Fax: (732) 449-4729. (703) 412-7500. what I know how to do”. I E-mail: wonder what the impact would [email protected] be if every attendee was called Call for to action in the same way? The International Journal of Quality and Reliability Memories About the Author: Management (IJQRM) is In December 1998, the IEEE Ellen Walker is a RAC seeking papers that focus on the specialist in software Reliability Society will practice of quality, reliability publish a special issue of the reliability. She holds and maintainability. IJQRM is Bachelor Degrees in IEEE Transactions on a major journal that will be Reliability commemorating the Mathematics and Computer entering its 15th year of Science and a Masters Degree in 50th anniversary of the publication. Authors should founding of the society. The Management. In a twelve-year submit three copies of the tenure as a computer scientist, issue will include a feature article to the North American presenting interesting she has worked with all Editor: phases of the software experiences of practitioners in Professor Christian N. Madu any of the assurance sciences. development cycle and Dept. of Management and supported both engineering Your funny, poignant or Management Science historically relevant memories services and business processes. Lubin School of Business She has been a facilitator and may be submitted at any time Pace University up to 1 August 1998. technical consultant for several 1 Pace Plaza long term quality initiatives, Reminiscences will not be New York, NY 10038 refereed, but will be edited to and is an Examiner for New Tel: (212) 346-1919 York State’s Excelsior (Quality fit the available space. Send Fax: (212) 346-1573. to the special issue editor: Awards) program. E-mail: [email protected] Anthony Coppola The 69th Shock and IITRI Vibration Symposium to be 201 Mill Street held October 12-16, 1998 in Rome NY 13440 Minneapolis/St. Paul, Tel: (315) 339-7075 Minnesota is seeking two Fax: (315) 337-9932 E-mail: [email protected]

RAC Journal, Volume 6, No. 1 9 New from RAC

New Catalog Available interfaces and help files at an expected price of Just released is the new RAC Catalog of Products $100 for U.S. orders and $120 for Non-U.S. orders. and Services. Besides presenting the latest listing Order code will be FMD-CD. Availability will be of RAC printed and software products and their announced in the RAC Journal. prices, the catalog provides descriptions and contacts for: Optoelectronics to be Next Component Applications Guide • RAC consulting services Subject Available in the Spring of 1998 will be Reliable • RAC training courses Applications of Optoelectronic Devices, a • The RAC Data Sharing Consortium compilation of information to assist optoelectronic equipment designers in enhancing the reliability of • Selected Topics in Assurance Related their products. It will provide criteria for selection Technologies (START), a set of introductory of devices, failure rate data, failure mode data, pamphlets available free from RAC potential reliability concerns and other information necessary to apply the part in a • RAC Journal subscriptions (also free) and reliable manner. The product will the latest in a advertising rates (not free) series of application guides (See Help From RAC, • The RAC website this issue). Ordering information will be presented in the RAC Journal when the publication is The catalog includes descriptions of products still in released. development. For example, computer text files for MIL-HDBK-338, Electronic Reliability Design New Software Tool Will Aid COTS Handbook, and for MIL-HDBK-470, Selection Maintainability Design Handbook, are described, A software tool, Selection of Equipment to Leverage priced, and labeled “available soon.” Commercial Technology (SELECT), will be offered by RAC to quantify the estimated reliability and The catalog is free on request to RAC. Use any of risk of using equipment designed for relatively the contact data listed on the back cover of this benign environments in applications where the

issue, or call (888) 722-8737. expected stresses Selection of (temperature, Equipment to LEverage Commercial C ommercial MilitaryMilitary vibration, shock, Equipment Failure Modes and Mechanisms Technologyechnology ApplicationsApplications Reference Updated humidity, etc.) In 1991, RAC published FMD-91, Failure Modes/ are considerably Mechanism Distributions, providing component more severe. The failure modes and their relative probabilities of tool was occurrence as an aid to fault tolerant design, the originally preparation of diagnostics, and failure modes, developed by IIT Research Institute (operator of effects and criticality analysis (FMECA). A new the RAC) under contract to Rome Laboratory (now document, FMD-97, adds about 50% new data to called Air Force Research Laboratory Information that of FMD-91. The hardcopy document contains Directorate - Rome Site). In addition to allowing data on electronic, mechanical and relative risk comparisons between different electromechanical parts and assemblies. It is commercial off-the-shelf (COTS) equipment, the available at a cost of $100 for U.S. orders, and $120 tool identifies the predominant environmental and for Non-U.S. orders. The order blank on the inside process risk drivers of each equipment being rear cover of this issue may be used to obtain a copy. considered, quantifying the impact of design For more information, call (888) 722-8737. changes and testing options on the risk scores and estimated reliability. Release is expected in FMD-97 will also soon be made available on CD- Spring, 1998. Order code will be SELECT, and the ROM or 3.5” diskette software packages for use on price will be $300 for U.S. orders and $340 for Non- personal computers under Windows 95 or NT. When U.S. orders. available, the software will include graphical user

10 RAC Journal, Volume 6, No. 1 RAC Journal, Volume 6, No. 1 11 Industry Brief

ISO 9000 in the News Recent news about ISO 9000, the family of Quality Management System international standards, includes:

• AS 9000, Aerospace Basic Quality System Standard, is now available from the Society of Automotive Engineers (SAE) as a guide to quality management in the Aerospace industry. Like QS 9000, the Automotive Quality System Standard, it contains a verbatim citation of ISO 9000 provisions and industry specific additional requirements and notes. For more information on AS 9000, Contact SAE, 400 Commonwealth Drive, Warrendale PA 15086-0001. Tel: (412) 776-4970.

• Despite recent statements that the automotive sponsors of QS 9000 would discontinue the verbatim citation of ISO 9001, present plans are to retain the ISO standard.

• ISO plans to “integrate” ISO 9000 and ISO 14000, Environmental Management System, apparently do not intend a merger of the documents, but rather the creation of the capability to audit a site for compliance to either or both in a single visit.

SLDA Coolers Developed Through SBIR Program Semiconductor laser diode arrays (SLDAs) require a high degree of thermal stability to maintain specified laser beam wavelengths and optimum efficiency. However, they generate large amounts of heat which must be removed, and existing cooling technology has been inadequate. Under a Small Business Innovation Research (SBIR) program sponsored by the former Air Force Phillips Laboratory, Saddleback Aerospace, Los Alamitos, CA, has developed water-cooled heat sinks using microchannel coolant flow passages . One of the prototype units exhibited the lowest thermal resistance ever reported for these types of devices. Saddleback is now manufacturing units in a variety of shapes, sizes and materials.

MTIAC Supporting Internet Product News Network The Manufacturing Technology Information Analysis Center (MTIAC), one of three Defense Technical Information Center (DTIC) information analysis centers operated by IIT Research Institute, is providing manufacturing expertise to assist the Thomas Magazine Group in the launch of it’s new commercial internet venture, the “Product News Network.” MTIAC is developing hierarchical nomenclatures for six categories of industrial products: adhesives and sealants; data acquisition; machine tools; quality control; automatic identification systems; and plant maintenance and equipment. The Product News Network is intended to be an on-line source of up-to-date product information. The Thomas Magazine Group is the publisher of the “Thomas Register of Manufacturers.”

ADPA + NSIA = NDIA On March 1, 1997, the American Defense Preparedness Association (ADPA) and the National Security Industrial Association (NSIA) merged. On October 1, 1997, the resulting organization was officially named The National Defense Industrial Association (NDIA). The association headquarters is located at 2111 Wilson Boulevard, Suite 400, Arlington VA 22201. Tel: (703) 522-1820. Fax: (703) 522-1885. E-mail: [email protected].

DTIC Offers IAC Directory The Defense Technical Information Center (DTIC) is offering a Directory of the DoD and service sponsored Information Analysis Centers (IACs). For a copy contact: Mr. Ron Hale, IAC Program Management Office, Defense Technical Information Center, 8725 John J. Kingman Road, Suite 0944, Ft. Belvoir VA 22060-6218. Tel: (703) 767-9171. Fax: (703) 767-9119. E-mail: [email protected]. The directory can also be downloaded from the IAC hub page on the DTIC webesite: http://www.dtic.mil/iac/. An Adobe reader is needed.

The appearance of advertising in this publication does not constitute endorsement by the Department of Defense or RAC of the products or services advertised.

12 RAC Journal, Volume 6, No. 1 From the Editor The Quality Troika

How often has been heard the complaint that an automaker might measure schedule and cost are the enemies of quality (or defect rate, sticker price and reliability, maintainability, testability, logistics time from order to delivery. planning, etc.)? I would like to suggest, rather, With this, or an equivalent set of three measures, that cost and schedule are appropriate and valid he could track his progress and benchmark his measures of quality. David Garvin once identified performance against the competition. These three five categories of quality measures, including parameters, and many others, are certainly used value-based measures and user-based measures. now by the automotive industry, but I am suggesting Value-based measures integrate “goodness” and the that a troika of the top three goodness, cost and cost involved (Phil Crosby’s “cost of quality “ schedule parameters may have value as an overall would be a value-based quality measure). User- figure of merit. Also, each component of an based measures try to quantify the ability to organization can select their own appropriate satisfy the customer. I submit all measures must troika as their metric for overall performance. And ultimately be user-based (no one else’s opinion the recommendations of the assurance specialists really counts) and that cost and schedule are should include a three-dimensional estimate of customer measures of quality, just as important as impact, using an appropriate quality troika. freedom from defects or features of the product. A characteristic of the quality troika is that any For example, when in the market for a car, would change which improves one leg without adversely you consider a Rolls-Royce? It has a reputation for affecting the other two is always a winner. Many unsurpassed engineering, elegant appointments, and process improvements fall into this category, and freedom from defects, making it one of the highest pay off handsomely. It is the case where an quality automobiles in conventional quality improvement in one leg hurts another that requires (“goodness”) terms. Yet the average person a trade-off analysis. In the past, the relatively considers the cost of an automobile important, and less visible “goodness” parameters suffered when the Rolls-Royce does not have the property of they crossed cost or schedule. In the current climate being priced low enough for consideration by most where quality (in terms of “goodness”) is car buyers. I contend this demonstrates cost is recognized as a customer demand, I think indeed a quality factor. management is more willing to listen. However, it’s up to the analyst to have his facts ready to How about schedule as a quality factor? If you clearly show the trade-off involved, or, better, to picked a new car to replace the one you have, but find a way where no leg of the troika suffers a loss had to wait for a year before it was delivered, (TQM, again). wouldn’t you pick another to buy? Perhaps you would wait, but I suspect you would think long and Considering cost and schedule as quality hard about it. QED: quality includes timely parameters, the assurance specialist may begin to delivery. understand why his recommendations have not always been accepted. More importantly, this may My conclusion is that measures of “goodness” join be a way for the specialist to communicate better with measures of cost and measures of schedule to with the decision makers, and it can improve the form a “quality troika,” the three legs of which value of his recommendations. Better must all be considered in making decisions. This is communication of better recommendations should essentially a premise of total quality management, improve his success ratio. (TQM), where the word “total” implicitly recognizes the existence of the troika. Anthony Coppola, Editor

So where does this lead? First, we need to measure What do you think? Letters to the editor on this, or our efforts in three dimensions. Among all the any other topic, are always welcome. Those of different parameters, we can perhaps select major general interest may be printed, unless the author factors for “goodness,” cost and schedule and use requests otherwise. this troika as an overall indicator of our performance. Continuing our automotive example,

RAC Journal, Volume 6, No. 1 13 14 RAC Journal, Volume 6, No. 1 RAC Journal, Volume 6, No. 1 15 be used in developing the Logistics Support Letter to the Editor Analysis for Maintenance Planning. It also provides the basis for Life-Cycle Cost Dear Sir, optimization, and is one of the most important Reference “FMEA - A Curse or Blessing” by S.J. building blocks in the development of a support Jakuba, in the RAC Journal, Volume 5, Number 4, I plan. would like to make the following comments: Nevertheless, in his appraisal of the design and The article was very well developed and written, development of systems he has put his finger on the and I feel it hit the right note regarding the most misused and misunderstood analysis, which if impressions the subject gives to both analysts, implemented and used in the manner he has designers, and other personnel normally exterior to suggested will benefit all areas of system/ reliability, who regard the usefulness of a FMEA equipment development. without any real understanding of its aims and achievability. Currently in the European environment most weapon platform acquisition programmes utilize a In discussing the uses of a FMEA Stan Jakuba does mixture of existing/modified/new development not address the uses of a FMEA when an item of equipments or a system comprising existing/ equipment or a system has to be modified or is an modified items with new software. With software existing item (off-the-shelf) where no previous and software/hardware configurations, existing FMEA has been developed or produced. In such FMEA methods do not suffice, but as Stan points out, cases one of the main uses of the FMEA is to define it is the manner in which the FMEA is approached, the types, numbers, frequencies and safety related the timing and how it is assessed, that is most hazard quantifications, so that the information can important, and which if implemented correctly will provide the most rewards. However, what must be clearly defined in each programme is why the FMEA is to be developed, and what are the aims of the FMEA. Most people involved in any system/equipment development programme fail to lift their heads out of the water and remind themselves what they are trying to achieve, and why they are doing such tasks. On the matter of training, Stan has it just right; experience, team effort, and teamwork are the only answer. Too often the team leader thinks that the rest of the team do the work and he/she is the team! Education is one possibility, but the real answers are far reaching, and go well beyond the subject of quality and reliability. Please, let’s have more articles like that of Stan Jakuba’s. The more people that read and digest such articles, the better are the chances of improving the quality of our programmes.

Yours very truly, Stewart Allen Logistics Executive Director, Philotech GmbH

16 RAC Journal, Volume 6, No. 1 Some Observations on Demonstrating Availability By: Anthony Coppola

Availability is easy to measure. It is also easy to on other distributions, such as the log-normal, measure MTBF and MTTR and calculate inherent which is more commonly assumed for MTTR). availability by the formula: Selected values from a Chi-square table are: MTBF Ai = MTBF+ MTTR Degrees of Chi-squared freedom ( = 2 x no. Chi-square value value at 90th of failures) at tenth percentile percentile The problem is in assigning risks. For example, 2 .211 4.61 suppose we wanted an availability of 95%. If we 4 1.064 7.78 had one failure in 100 hours and it took one hour to 6 2.20 10.64 8 3.49 13.36 repair, we have measured an availability of 99%, 10 4.87 15.99 well over spec. However, we would be basing the conclusion on one failure and one repair, which 12 6.30 18.55 14 7.79 21.06 would be quite risky. We would feel more 16 9.31 23.54 confidence if we had more data (i.e. more failures 18 10.86 25.99 and repairs). There are established procedures for 20 12.44 28.41 computing confidence intervals around MTBF 22 14.04 30.81 measurements and for confidence intervals around 24 15.66 33.20 MTTR estimates. (confidence = 1-risk. A confidence 26 17.29 35.56 28 18.94 37.92 of 90% that a calculated MTBF or better has been 30 20.60 40.26 achieved means a 10% risk of being in error. In turn a 10% risk of error means a probability of 0.10 that 40 29.05 51.81 50 37.69 63.17 the true MTBF will be lower than the calculated 60 46.46 74.40 value.) As might be expected, the more data 80 64.28 96.58 available, the closer will the calculated values for 100 82.36 118.5 a specified confidence be to the measured value. We then use the formulas: The problem with availability is that there is no 2 x total operating time easy way to compute confidence limits about a MTBF = measured availability. Chi - square value for 2 x no. of failures at 90th percentile Following is a procedure for calculating risks on availability from confidence intervals on MTBF 2 x total downtime and MTTR. MTTR = Chi - square value for 2 x no. of First, collect data on MTBF and MTTR. (MTBF is repairs at tenth percentile operating hours/failures: MTTR is down time/failures not counting administrative and The MTBF formula gives the value of MTBF which logistics delays). The data is collected to the end we are 90% sure will be exceeded by the true MTBF of the last repair so that the operating time does (there is a .10 probability that the true MTBF is not extend beyond the time of the last failure. less that that calculated by the formula).

Then use the Chi-square procedure described below The MTTR formula gives the value of MTTR which to calculate MTBF and MTTR at a desired risk. We we are 90% sure will not be exceeded by the true shall be using 10% risks, and calculating a value of MTTR (there is a .10 probability that the true MTBF for which there is only a .10 probability MTTR is grater than that calculated by the that the true MTBF will be lower, and a value of formula). MTTR for which there is only a .10 probability that the true MTTR will be higher. The resultant These values may be used to calculate an values are an MTBF and MTTR demonstrated to a availability at some risk determined by the risks 90% confidence. (Note: this method assumes the on MTBF and MTTR. We shall discuss later what exponential distribution applies to both times the risk on the calculated availability might be. between failure and times to repair: there are Example: for 1000 hours of operating time with 10 procedures for computing confidence intervals based failures which required 10 hours of repair time:

RAC Journal, Volume 6, No. 1 17 1000 calculated is also .10. Hence the probability of Measured MTBF = = 100 hours; 10 both values being worse is (.10 x .10 = .01). In such a 10 case, the true availability must be worse than the Measured MTTR = = 1 hour. calculated value using the calculated MTBF and 10 MTTR. Hence the risk on the calculated Using the formulas given above: availability is at least 1% (we shall show it is more) and the confidence (1 - risk) cannot exceed 2000 99% MTBF = = 70.4 hours 28. 41 The above cases cover 82% of all possibilities. This 20 leaves 18% probability that the true value of one of MTTR = = 1.6 hours 12 . 44 the figures of merit (MTBF or MTTR) is better than the calculated value and the true value of the The results are interpreted as showing only a 10% other is worse. In this case the true value of chance that the true MTBF will be less than 70.4 availability may be either better or worse than the hours, and that there is only a 10% chance that the calculated availability. Unfortunately, the true MTTR will exceed 1.6 hours. majority of the time, the true availability will be worse, rather than better than that calculated. Calculating availability by the formula A = This is because a slight deviation in the percentile MTBF/(MTBF+MTTR): towards the tails of the distribution (the bad way) makes a bigger difference (in calculating MTBF or 100 100 Measured availability = = = .99 MTTR) than the same deviation towards the center 100 + 1 101 (the good way). For example, The difference in the 90th and 91th percentile values is greater than the Availability using calculated MTBF and MTTR: difference between the 90th and 89th percentiles. Hence, in less than half the cases will the true 70. 4 70. 4 A = = = .978 availability exceed the calculated value. So the + 70.. 4 1 6 72 contribution of this 18% probable situation to the confidence in the calculated value will be less than If the desired availability were .98, the measured 9%. The exact value will depend on how sharp the results would indicate achievement, but the curves are skewed and the ratio of MTBF to MTTR. calculated results would show that when the MTBF and MTTR are calculated to a 90% confidence, the We can conclude that calculating the availability availability is not demonstrated as being using the 90% confidence values of MTBF and MTTR achieved. More test data would be needed to will provide a confidence of something more than resolve the issue. 81% but less than 90% (81% + less than 9% as shown above) that the true availability will The important, as yet unanswered, question is: exceed the calculated value. This will often be what is the risk on the calculated availability satisfactory. number? Using other values of confidence (e.g., 95%) for the In attempting to answer the question, note that the calculation of MTBF and MTTR will result in other calculated MTBF will be exceeded by the true ranges of confidence on availability. The above MTBF with probability .90 and the true MTTR will procedure, and a more complete chi-square table be less than the calculated MTTR with probability will be needed. Note however, that the higher the .90. Hence, the probability of both figures of merit confidence demanded, the longer the test time being better than calculated is .90 x .90 = .81. When needed to achieve it. Also note that all this both figures of merit are better than calculated, the assumes that the measured availability (that true availability must exceed the value calculated calculated from the measured MTBF and MTTR) from the calculated MTBF and MTTR. So our exceeds the specified availability. If the confidence in the calculated availability (the measured availability is inadequate, the probability that the true availability will be confidence in any higher value must be less than greater than the calculated value) is at least 81%. 50% (a poor bet). (we shall show that it is more). Note that the probability of the true MTBF being worse than that calculated is .10 and the probability of the true MTTR being worse than that

18 RAC Journal, Volume 6, No. 1 RAC Journal, Volume 6, No. 1 19 Mark Your Calendar

May 18-22, 1998 Tuscon, AZ 0119. Tel: (520) 621-6120. Fax: (520) 621-8191. E-mail: 24th Annual Reliability Testing Institute [email protected]. Contact: Dr. Kececioglu, University of Arizona, Aerospace and Mechanical Eng. Dept., 1130 N. Mountain Avenue, August 26-28, 1998 Albuquerque, NM Bldg. 119, Room N517, PO Box 210119, Tuscon, AZ 85721- 1998 Military/Aerospace (Transportation) COTS 0119. Tel: (520) 621-6120. Fax: (520) 621-8191. E-mail: [email protected]. Conference Contact: Edward B. Hakim, The Center for Commercial Component Insertion, Inc., 2412 Emerson Ave., Spring Lake, May 19-21, 1998 Adelaide, Australia NJ 07762. Tel: (732) 449-4729. E-mail: Maintenance Engineering ‘98 [email protected] Contact: Secretariat, Maintenance Engineering ‘98, PO Box 5142, Clayton Victoria 3168, Australia. Tel: 61 3 9544 September 13-17, 1998 New York, NY 0066. Fax: 61 3 9543 5905. International Conference on Probabilistic Safety Assessment and Management June 4, 1998 Pittsburgh. PA Contact: Dr. R. A. Bari, Brookhaven National Laboratory, November 5, 1998 Arlington, VA PO Box 5000, Upton, NY 11973-5000. tel: (516) 344-5266. SEI Visitor’s Days Fax: (516) 344-5266. E-mail: [email protected]. Contact: Customer Relations, Software Engineering Institute (SEI), Carnegie Mellon University, Pittsburgh, PA September 14-19, 1998 Seattle, WA 15213-3890. Tel: (412) 268-5800. E-mail: customer- [email protected]. More info at website at URL: 16th International System Safety Conference http://www.sei.cmu.edu. Contact: Clif Ericson, 18247 150th Ave SE, Renton, WA 98058. Tel: (253) 657-5245. Fax: (253) 657-2585. E-mail: [email protected] June 8-10, 1998 Crystal City, VA SEI Conference on Risk Management September 22-25, 1998 Ypsilanti, MI Contact: Customer Relations, Software Engineering Practical Implementation of Probabilistic Methods Institute (SEI), Carnegie Mellon University, Pittsburgh, PA 15213-3890. Tel: (412) 268-5800. E-mail: customer- Workshop [email protected]. More info at website at URL: Contact: SAE Professional Development, 400 http://www.sei.cmu.edu. Commonwealth Drive, Warrendale, PA 15096-0001. Tel: (724) 772-7148. Fax: (724) 776-4955. E-mail: [email protected]. June 15-17, 1998 Dallas, TX 10th Annual RMSL October 6-8, 1998 Reno, NV Reliability, Maintainability, Supportability, and 20th Annual EOS/ESD Symposium Logistics Workshop Contact: ESD Association, 7900 Turin Rd., Bldg. 3, Suite 2, Contact: Professional Development Division, SAE, 400 Rome, NY 13440-2069. Tel: (315) 339-6937. Commonwealth Drive, Warrendale PA 15096-0001. Tel: (724) 772-7148. Fax: (724) 776-4955. E-mail: [email protected]. October 26-29, 1998 Annapolis, MD 20th Space Simulation Conference Contact: Institute of Environmental Sciences & June 23-25, 1998 Munich, Germany Technology, 940 East Northwest Highway, Mount FTCS-28: Fault Tolerant Computing Symposium Prospect, IL 60056. Tel: (847) 255-1561. Fax: (847) 255- Contact: Ram Chillarege, IBM Watson Research, 30 Saw 1699. E-mail: [email protected]. Mill River Road, Hawthorne, NY 10352. Tel: (914) 784- 7375. Fax: (914) 784-6267. E-mail: [email protected] November 5, 1998 Arlington, VA SEI Visitor’s Day July 15-17, 1998 Tokyo, Japan Contact: Software Engineering Institute (SEI), Carneige 5th ISPE International Conference on Concurrent Mellon University, Pittsburg, PA 15213-3890. Tel: (412) Engineering 268-5800. E-mail: [email protected]. Contact: Prof. Shuichi Fukuda, Dept. of Production, Information and Systems Engineering, Tokyo Metropolitan November 16-20, 1998 Tucson, AZ Institute of Technology, 6-6, Asahigaoka, Hino, Tokyo 191, 36th Annual Reliability Engineering & Japan. Tel: 81 425 83 5111 x3605. Fax: 81 425 83 5119. E-mail: [email protected]. Management Institute Contact: University of Arizona, 1130 N. Mountain Ave., Tucson, AZ 85721-0119. Tel: (520) 621-6120. Fax: (520) July 27-30, 1998 San Diego, CA 621-8191. E-mail: [email protected]. Seventh Applied Reliability Engineering and Product Assurance Institute for Engineers and Managers Contact: Dr. Kececioglu, University of Arizona, Aerospace and Mechanical Eng. Dept., 1130 N. Mountain Avenue, Bldg. 119, Room N517, PO Box 210119, Tuscon, AZ 85721-

20 RAC Journal, Volume 6, No. 1 RAC Journal, Volume 6, No. 1 21 Help from RAC

Complementing the RAC may be successfully exploited. handbook, Processes for product in process on the Order Code: MCM Using Commercial Parts in reliable application of Price: $50 U.S., $60 Non-U.S. Military Applications. The optoelectronics devices (see handbook provides an New From RAC, this issue) are Reliable Application of overview of the acquisition several currently available Capacitors provides reform process, its relationship publications on applying other information on the failure to system engineering functions, part types and on general parts modes of the and its impact on reliability management. Descriptions various and quality issues. It also follow: Reliable capacitor types, Application provides guidance for a parts their observed of Capacitors management program under the

Reliable Application of failure rates and RAC Parts Selection, Application and Control S new groundrules, and case Plastic Encapsulated specific areas of C histories to illustrate practical ES R Microcircuits addresses the concern when LL R L experience and “lessons R interest in applying plastic selecting a I learned.” encapsulated microcircuits capacitor. It Order Code: PUCP Reliability Analysis Center

RAC is a DoD Information Analysis Center (PEMs) in all environments also contains Sponsored by the Defense Technical Information Center Price: $25 U.S., $35 Non-U.S. including military and quantitative relationships telecommunication systems describing the change in Orders may be placed using the with strict reliability demands electrical parameters order blank on the following and harsh environments. This (capacitance, dissipation factor page, or by telephone to (888) interest stems from real and and equivalent series 722-8737. Other RAC products perceived improvements in resistance) as a function of time are listed in the RAC Product PEM reliability, quality and and temperature. A thorough and Services Catalog, recently cost. The bibliography is provided. revised. Ask for your free copy. Reliable publication Application Order Code: CAP provides of Plastic Price: $50 U.S., $60 Non-U.S. Encapsulated information on Microcircuits RAC on the Web PEM parts RAC Parts Selection, Application and Control Seri The first publication in RAC’s control, series on the reliable RAC maintains a home page on procurement, application of parts was Parts the World Wide Web at URL: failure modes Selection, Application and http://rome.iitri.com/RAC/. Reliability Analysis Center

RAC is a DoD Information Analysis Center and Sponsored by the Defense Technical Information Center Control, an overview From there, the user can obtain: mechanisms, reliability and document addressing topics of a quality. A reliability • Current awareness general nature affecting all information: copies of the prediction model for PEMs is types of parts. It identifies RAC Journal, START sheets, also provided. specific selection and a calendar of events Order Code: PEM2 • RAC products and training procurement methods and tools information: catalog of Price: $75 U.S., $85 Non-U.S. for properly applying products, on-line order form, electrical and electronic parts training schedules, available Reliable Application of courses most suitable for the unique • RAC services: RAC user Multichip Modules requirements of a given guide, bibliographic addresses hybrid microcircuits, application, military or searches, technical support, multichip packaging and databases commercial. • Links to other DoD multichip modules (MCMs). It Order Code: PSAC Information Analysis Centers provides an overview of the Price: $50 U.S., $60 Non-U.S. (IACs), and other reliability, methodology and tools to ensure maintainability and quality information sources that these components and To help bridge the gap between their vendors are selected, the use of military Pay us a visit at: controlled and tested specifications and standards, effectively for the specific end and new policies in acquiring http://rome.iitri.com/RAC/ product and usage, so that the military equipment under the favorable weight and size DoD acquisition reform advantages of these devices program, RAC compiled a

22 RAC Journal, Volume 6, No. 1 RAC Order Form

Send to: Reliability Analysis Center, 201 Mill Street, Rome, NY 13440-6916

Call: (888) RAC-USER (722-8737), (315) 339-7047 Fax: (315) 337-9932

E-mail: [email protected]

Qty Title Price Each Total

Shipping & Handling: US Orders add $4.00 per book for first class shipments, ($2.00 for RAC Blueprints) Non-US add $10.00 per book for surface mail (8-10 weeks), $15.00 per book for air mail ($25.00 for NPRD and VZAP, $40.00 for EPRD). ($4.00 for RAC Blueprints) Total

Name

Company

Division

Address

City State Zip

Country Phone Ext.

Fax: E-mail:

Method of Payment:

❒ Personal check enclosed ❒ Company check enclosed (make checks payable to IITRI/RAC) ❒ Credit Card # Expiration Date Type (circle): AMERICAN EXPRESS VISA MASTERCARD A minimum of $25 is required for credit card orders.

Name on card:

Signature:

Billing address:

❒ DD1155 (Government personnel) ❒ Company purchase order

❒ Send product catalog

RAC Journal, Volume 6, No. 1 23 RAC Training Courses

Course Dates: Design Reliability Bring RAC Training Courses This intensive overview covers theoretical to Your Facility June 9-11, 1998 and practical aspects of reliability through RAC provides training courses to concurrent engineering. Reliability analysis, government and industry in virtually every Course Fee: test and evaluation, parts selection, circuit aspect of reliability and quality. On-site analysis, and applicable standards and training is often more cost-effective for $995 information sources are addressed. organizations, particularly since on-site “closed” courses can be tailored to specific Location and On-Site Mechanical Reliability customer products and processes. Subjects Practical applications of mechanical addressed by RAC training include but are Accommodations: engineering to system and component not limited to: reliability are covered in this popular • Design for Reliability Virginia Beach Resort Hotel & course. Basic theories of mechanical • Reliability Modeling Conference Center reliability and the essential tools of • Fault Tree Analysis mechanical reliability analysis are • Failure Analysis 2800 Shore Drive • Statistical Process Control covered and reinforced through • Software Engineering Virginia Beach, VA 13451 problem solving and discussion. • Software Reliability (757) 481-9000 • Failure Data Systems • Mechanical Reliability Accelerated Testing • Reliability Testing The course describes the statistical models • Testability Analysis System Software Reliability for accelerated tests, how to plan efficient • Reliability Analysis Featuring hands-on software reliability tests, and how to estimate and improve • Reliability Management/Planning measurement, analyses and design, this product reliability. The methods will be • Quantitative Methods course is intended for those responsible for illustrated with many applications to • Microelectronics Standardization measuring, analyzing, designing, electronic, mechanical, and other products, • Worst Case Circuit Analysis • Maintainability Testing automating, implementing, or ensuring including your own data, which you are • Failure Mode, Effects & Criticality Analysis software reliability for commercial or encouraged to bring. • Total Quality Management government programs. • Environmental Stress Screening • Parts Selection and Control • Reliability-Centered Maintenance • Probabilistic Mechanical Design • Qualified Manufacturers List

For further information on scheduled RAC training courses contact Ms. Nan Pfrimmer at the Reliability Analysis Center, (800) 526-4803 or (315) 339-7036. For information about on-site and custom training arrangements contact Mr. Patrick Hetherington at (315) 339-7084..

IIT Research Institute/ Non-Profit Org. Reliability Analysis Center U.S. Postage 201 Mill Street Paid Utica, NY Rome, NY 13440-6916 Permit No. 566

RAC INQUIRIES AND ORDERS:

Reliability Analysis Center 201 Mill Street Rome, NY USA 13440-6916

(315) 337-0900 ...... Main Telephone (315) 337-9932 ...... Facsimile (315) 337-9933 ...... Technical Inquiries (800) 526-4802 ...... Publication Orders (800) 526-4803 ...... Training Course Info (888) RAC-USER (722-8737) ..... General Inquiries [email protected]...... E-mail (Product Info) [email protected] ...... E-mail Technical Inquiries URL: http://rome.iitri.com/RAC ...... Web Address Contact Gina Nash at (800) 526-4802 or the above address for our latest Product Catalog or a free copy of the RAC Users Guide.