PREFACE

Reliability is one of the most important quality characteristics of components, products, and large and complex systems. The role of reliability is observed daily by each one of us, when we start a vehicle, attempt to place a phone call, use a copier, use a computer, or take a train. In all instances, the user expects that the machine or the system to provide the designed functions as expected. Most likely, you have experienced machines that do not always function or deliver the desired quality of service when needed. Machines and systems experience failures, interruption, and possibly termination of service.

Engineers spend a significant amount of time and resources during the design, product (or service) development, and production phases of the product life cycle to ensure that the product or system will provide the desired service level. In doing so, engineers start with a concept design, select its components, construct prototypes, test its functionality, and estimate its reliability. Modifications and design changes are usually made, and these steps are repeated until the product (or service) satisfies its reliability requirements. The prelude of this book presents these steps in the design and life cycle of the "One-Hoss-Shay."

Designing the product may require redundancy of components (or subsystems), or introduction of newly developed components or materials or changes in design configuration. These will have a major impact on the product reliability. Once the product is launched and used in the field, data are collected, so improvements can be made in the newer versions of the product. Moreover, these data become important in identifying potential safety issues or hazards for the users, so recalls can be quickly made to resolve these issues. In other words, reliability is a major concern during the entire life of the product and is subject to continuous improvement. This is noticeable by the frequent and continuous updates of the operating systems of cell phones, computers, and software applications.

This book is an engineering reliability book. It is organized according to the same sequence followed when designing a product or service. The book consists of four parts. Part I focuses on system reliability estimation for time-independent and time-dependent models. Chapter 1 addresses on the basic definitions of reliability, extensive coverage of failure-time distributions and their hazard functions, reliability metrics, and methods for its calculations. Chapter 2 describes, in greater detail, methods for estimating reliabilities of a variety of engineering systems configurations starting with series systems, parallel systems, series-parallel, parallel-series, consecutive k-out-of-n:F, k-out-of-n, and complex network systems. It also addresses systems with multistate devices and concludes by estimating reliabilities of redundant systems and the optimal allocation of components in a redundant system. Finally, several importance measures of components in the system are presented since these measures could be used to determine the components and subsystems that require "hardening" through replacements of components with "more reliable" ones or assigning higher priorities of repair in case of failures. The next step in the product design is to study the effect of time on system reliability, since reliability is a time-dependent characteristic of the products and systems. Therefore, Chapter 3 discusses, in detail, time- and failure-dependent reliability and the calculation of mean time to failure (MTTF) of a variety of system configurations. It also introduces availability as a measure of system reliability for repairable systems. Once the design is "firm," the engineer assembles the components and configures them to achieve the desired reliability objectives. This may require conducting reliability tests on components or using field data from similar components.

Therefore, Part II of the book, starting with Chapter 4, presents introduces for estimating the parameters of the failure-time distributions including method of moments, regression, and the concept of constructing the likelihood function and its use in estimating the parameters. Chapter 5 provides a comprehensive coverage of parametric and nonparametric reliability models for failure data (censored or noncensored) and testing for abnormally long or short failure times. The extensive examples and methodologies, presented in this chapter, will aid the engineer in appropriately modeling the test data. Confidence intervals for the parameters of the models are also discussed. More importantly, the book devotes a full chapter, Chapter 6, to accelerated life testing and degradation testing. The main objective of this chapter is to provide varieties of statistical-based models, physics-statistics-based models, and physics-experimental-based models to relate the failure time and data at accelerated conditions to the normal operating conditions at which the product is expected to operate.

This leads to Part III, which focuses on the understanding of failure causes, mechanism of failures, and the physics of failures, as described in Chapter 7. This chapter also provides the physics of failure of the failure mechanisms in electronic and mechanical components. It demonstrates the use of the parameters of the failure mechanism in the estimation of the reliability metrics. In addition to making a system reliable, other metrics may include resilience that demonstrates the ability of the system to absorb and withstand different hazards and threats. This is detailed in Chapter 8, where resilience quantifications of both nonrepairable and repairable systems are presented and demonstrated through examples.

Finally, once a product is produced and sold, the manufacturer and designer must ensure its reliability objectives by providing preventive and scheduled maintenance and warranty policies. Part IV of the book focuses on these topics; it begins with Chapter 9, which presents different methods (exact and approximate) for estimating the expected number of system failures during a specified time interval. These estimates are used in Chapter 10 in order to determine optimal maintenance schedules and optimum inspection policies. Methods for estimating the inventory levels of spares required to ensure predetermined reliability and availability values are also presented. Chapter 11 explains different warranty policies and approaches for determining the product price including warranty cost, as well as, the estimation of the warranty reserve fund. Chapter 12 concludes the book. It presents actual case studies which demonstrate the use of the approaches and methodologies discussed throughout the book in solving real life cases. The role of reliability during the design phase of a product or a system is particularly emphasized.

Every theoretical development in this book is followed by an engineering example to illustrate its application. In addition, many problems are included at the end of each chapter. These two features increase the usefulness of this book as being a comprehensive reference for practitioners and professionals in the quality and reliability engineering area. In addition, this book may be used for either a one- or two-semester course in reliability engineering geared toward senior undergraduates or graduate students in industrial and systems engineering, mechanical, and electrical engineering programs. It may also be adapted for use in a life data analysis course offered in many graduate programs in statistics. The book presumes a background in statistics and probability theory and differential calculus.

ACKNOWLEDGMENTS

This book represents the work of not just the author, but also many others whose works are referenced throughout the book. I have tried to give adequate credit to all those whose work has influenced this book. Particular acknowledgments are made to the Institute of Electrical and Electronic Engineers, CRC Press, Institute of Mathematical Statistics, American Society of Mechanical Engineers, Seimens AG, Electronic Products, and Elsevier Applied Science Publishers for the use of figures, tables in the appendices, and permissions to include material in this book.

I would like to thank the students of the Department of Industrial and Systems Engineering at Rutgers University, who have been using the earlier versions and editions of this book for the past 25 years and provided me with valuable input. In particular, I would like to thank Askhat Turlybayev, Xi Chen, and Yao Cheng for providing extensive input and comments and Changxi Wang, for drawing many of the figures and providing the corrosion example in Chapter 7. I also acknowledge the collaboration with Haitao Liao and Xiao Liu of the University of Arkansas.

Special gratitude goes to the Council for International Exchange of Scholars for the Fulbright Scholar and to the National Science Foundation, the Federal Aviation Administration, and many industries who supported my research and teachings over many years. Special thanks to one of my favorite students, John Sharkey and his wife Chris, for their generous support that provided me with release time to complete this book.

I would like to acknowledge Dr. Mohammed Ettouney for his support and many detailed discussions regarding resilience and failure examples of civil engineering infrastructures. I am also thankful to Joe Lippencott for...