Reliability Engineering

Considering that complex equipment and systems are generally repairable, contain redundancy and must be safe, the term reliability appears often for reliability, availability, maintainability, and safety. RAMS is used to point out this wherever necessary in this book. The purpose of reliability engineering is to develop methods and tools to assess RAMS figures of components, equipment and systems, as well as to support development and production engineers in building in these characteristics. In order to be cost and time effective, reliability (RAMS) engineering must be integrated in the project activities, support quality assurance and concurrent engineering efforts, and be performed without bureaucracy. This chapter introduces basic concepts, shows their relationships, and discusses the tasks necessary to assure quality and reliability (RAMS) of complex equipment and systems with high quality and reliability (RAMS) requirements. A comprehensive list of definitions with comments is given in Appendix A1. Standards for quality and reliability (RAMS) assurance are discussed in Appendix A2. Refinements of management aspects are in Appendices A3 - A5. Risk management is considered in Sections 1.2.7 and 6.11.

Reliability analysis during the design and development of complex components, equipment and systems is important to detect and eliminate reliability weaknesses as early as possible and to perform comparative studies. Such an investigation includes failure rate and failure mode analysis, verification of the adherence to design guidelines, and cooperation in design reviews. This chapter presents method and tools for failure rate and failure mode analysis of complex equipment and systems considered as nonrepairable up to system failure (except for Eq. (2.48)). Estimation and demonstration of a constant failure rate λ or of MTBF for the case MTBF ≡ 1/λ is in Section 7.2.3. After a short introduction, Section 2.2 deals with series - parallel structures. Complex structures, elements with more than one failure mode, and parallel models with load sharing are investigated in Section 2.3. Reliability allocation with cost considerations is discussed in Section 2.4, stress / strength and drift analysis in Section 2.5. Section 2.6 deals with failure mode and causes-to-effects analyses, and Section 2.7 gives a checklist for reliability aspects in design reviews. Maintainability is considered in Chapter 4, together with spare parts reservation and maintenance strategies with cost considerations. Repairable systems are investigated in Chapter 6 including complex systems for which a reliability block diagram does not exist, imperfect switching, incomplete coverage, reconfigurable systems, common cause failures, as well as an introduction to network reliability, BDD, ET, dynamic F T, Petri nets, computer-aided analysis, and human reliability. Risk management for repairable systems is considered in Section 6.11. Design guidelines are given in Chapter 5, qualification tests in Chapter 3, reliability tests in Chapters 7 and 8. Theoretical foundations for this chapter are in Appendix A6.

Materials, components, and assemblies have a great impact on the quality and reliability of the equipment and systems in which they are used. Their selection and qualification has to be considered with care by new technologies or important redesigns, on a case-by-case basis. Besides cost and availability on the market, important selection criteria are intended application, technology, quality, longterm behavior of relevant parameters, and reliability. A qualification test includes characterization at different stresses (for instance, electrical and thermal for electronic components), environmental tests, reliability tests, and failure analysis. After some considerations on selection criteria for electronic components (Section 3.1), this chapter deals with qualification tests for complex integrated circuits (Section 3.2) and electronic assemblies (Section 3.4), and discusses basic aspects of failure modes, mechanisms, and analysis of electronic components (Section 3.3). Procedures given in this chapter can be extended to nonelectronic components and materials as well. Reliability related basic technological properties of electronic components are summarized in Appendix A10. Statistical tests are in Chapter 7, test and screening strategies in Chapter 8, design guidelines in Chapter 5.

At equipment and systems level, maintainability has a great influence on reliability and availability. This holds, in particular, if redundancy has been implemented and redundant parts can be repaired (restored) on-line, i. e. without operation interruption at system level. Maintainability is thus an important parameter in the optimization of reliability, availability, and life-cycle cost. Achieving high maintainability in complex equipment and systems requires appropriate activities which must be started early in the design and development phase and be coordinated by a maintenance concept. To this belong partitioning of equipment and systems into (as far as possible) independent line replaceable units (LRU's), failure detection and localization (built-in tests), and logistic support. A maintenance concept has to be tailored to the equipment or system considered. Its definition and realization must be actively supported by the project manager. After some basic concepts, Section 4.2 deals with a maintenance concept for complex equipment and systems. Section 4.3 discusses maintainability aspects in design reviews. Section 4.4 gives methods and tools for maintainability prediction. Spare parts provisioning and repair strategies are carefully considered in Sections 4.5 and 4.6, cost optimization in Sections 4.5 - 4.7. Design guidelines for maintainability are given in Section 5.2. The influence of preventive maintenance, imperfect switching and incomplete coverage, as well as reward and frequency / duration aspects are investigated in Section 6.8. For simplicity, delays (administrative, logistic, technical) are neglected and repair is used for restoration.

Reliability, maintainability, safety, and software quality have to be built into complex equipment and systems during the design and development phase. This has to be supported by analytical investigations (Chapters 2, 4, 6) as well as by design guidelines and tests (Chapters 5, 3, 7, 8). Developing design guidelines demands practical experience and engineering feeling. Adherence to such guidelines limits the influence of those aspects which can invalidate the models assumed for analytical investigations, and improve the inherent reliability, maintainability, and safety of both hardware and software. Each industry producing equipment and systems with high reliability (RAMS) requirements is aware of the necessity for such guidelines. This chapter gives a comprehensive list of design guidelines for reliability, maintainability (incl. human and safety aspects), and software quality of complex electronic and electromechanical equipment and systems, harmonized with industry's needs, in particular for military and space applications).

Reliability and availability analysis of repairable systems is generally performed using stochastic processes, including Markov, semi-Markov and semi-regenerative processes (Table 6.1). A careful introduction to these processes is in Appendix A7 with reliability applications in mind. Equations for Markov and semi-Markov models are summarized in Table 6.2. This chapter investigates many of reliability models encountered in practical applications (some of which new in Sections 6.8 - 6.11), and

reliability figures at system level will have indices S i. (e. g. MTTF _Si ), where S stands for system and i for the state Z _i entered at t = 0 (footnote on p. 512).

After a discussion on assumptions and conclusions, Section 6.2 investigates the one item structure under general conditions. Sections 6.3 - 6.6 deal with series, parallel and series - parallel structures. To unify models and simplify calculations, it is assumed that systems have only one repair crew and no further failures occur at system down. Starting from constant failure and repair rates (Markov models) generalization is performed beginning with repair rates, up to the case in which the process involved is regenerative with a minimum number of regeneration states. Approximate expressions for large series - parallel structures are given in Section 6.7. Section 6.8 considers systems with complex structure for which a reliability block diagram often does not exist. On the basis of practical examples, preventive maintenance, imperfect switching, incomplete coverage, elements with > 2 states, phased-mission systems, common cause failures, and general reconfigurable fault tolerant systems with reward and frequency / duration aspects are investigated. Network reliability is introduced in Section 6.8.8 and a general procedure for complex structures is in Section 6.8.9. Section 6.9 considers alternative methods (dynamic FTA, BDD, event trees, Petri nets, computer-analysis), and gives a Monte Carlo approach useful for rare events. Sections 6.10 and 6.11 deal with human reliability and risk management. Results are summarized in tables. Asymptotic and steady-state is used for stationary, mean for expected value. Selected examples illustrate the practical aspects.

Statistical quality control and reliability (RAMS) tests are performed to estimate or demonstrate quality and reliability characteristics on the basis of data collected from sampling tests. Estimation leads to point or interval estimate, marked with ^ in this book; demonstration is a test of a given hypothesis on an unknown characteristic. Estimation and demonstration of an unknown probability is given in Section 7.1 for a defective probability p and in Section 7.2.1 for some reliability figures. Procedures for steady-state availability estimation and demonstration are in Section 7.2.2. Estimation and demonstration of a constant failure rate λ (or MTBF for MTBF = 1/λ) are discussed in depth in Sections 7.2.3. The case of an MTTR is considered in Section 7.3. Basic models for accelerated tests are introduced in Section 7.4, with regard also to multiple failure mechanisms. Goodness-of-fit tests based on graphical and analytical procedures are summarized in Section 7.5. General reliability data analysis, including test on nonhomogeneous Poisson processes and trend tests, are discussed in Section 7.6; models for reliability growth in Section 7.7. A careful introduction to the mathematical foundations for this chapter is in Appendix A8. Selected examples illustrate the practical aspects, and

to simplify the notation, sample is used for random sample (taken from a very large, homogeneous lot), mean for expected value and independent for mutually, statistically, stochastically independent; furthermore, indices S _i (p. 3) are omitted in this chapter ( S ₀ is assumed for R, R(t), PA, λ, MTBF), and estimated (or empirical) values are marked with ˆ .

Reliability (RAMS) assurance has to be continued during the production phase, coordinated with other quality assurance activities. In particular, for monitoring and controlling production processes, item configuration, in-process and final tests, screening procedures, and collection, analysis and correction of defects and failures. The last measure yields to a learning process whose purpose is to optimize the quality of manufacture, taking into account cost and time schedule limitations. This chapter introduces some basic aspects of quality and reliability (RAMS) assurance during production, discusses test and screening procedures for electronic components and assemblies, introduces the concept of cost optimization related to a test strategy and develops it for a cost optimized test and screening strategy at the incoming inspection. For greater details one may refer to [7.1 - 7.5, 8.1 - 8.14] for qualification and monitoring of production processes, as well as to [8.21 - 8.35] for test and screening aspects. Models for reliability growth are discussed in Section 7.7.