As Meng (2009) made clear, one of the statistics profession’s responsibilities is to be “the first quantitative trainers of future generations of scientists, engineers, policy makers, etc.” (not just statisticians). Evidence suggests we have not met this challenge. In fact, our traditional Stat101 courses and texts can poison the statistical well for the people who become our potential sponsors and collaborators. We need to do more than teach ‘methods.’ We need to show from the first day and throughout the Stat101 experience that our methods exist to help people learn interesting things about issues and topics they are passionate about. This message pertains to the rising generations of professionals and the citizenry at large and it applies to statisticians. Getting the message across may require radically redesigned ‘service courses’ and a new generation of uber-teachers as Meng (2009) advocated. In the meantime we should use existing materials in ways that show how subject-matter passion can motivate statistical analyses that reveal interesting and important subject-matter insights. As we develop new texts and other materials we need better quality control by authors, editors, and reviewers to assure that our teaching supports our “first quantitative trainer” responsibility.

This book presents new approaches for setting quantitative reliability requirements based on the cost of failure and specified minimum failure-free operation periods (MFFOPs). The cost-of-failure–based reliability analysis provides a real alternative to the current reliability analysis approach by considering the cost of the failures when setting reliability requirements. After an introduction to the reliability and risk analysis based on random variables, the author examines a new method for problem solving in the context of real reliability engineering applications and case studies and supplies algorithms that can be used for reliability analyses and for setting quantitative reliability requirements. The author also gives a comprehensive overview of basic Monte Carlo simulation techniques and algorithms for solving reliability engineering problems. In addition, the book provides a comprehensive introduction to loadstrength interference models for reliability and risk analysis by introducing the overstress reliability integral, a generalization of the load-strength interference integral with the time included. It also presents an efficient model for determining the probability of failure of loaded components and structures with integral flaws. The book comprises 16 chapters. Chapter 1 provides a general introduction to reliability (survival) functions, cumulative distributions and probability density functions of times to failure, and random events in reliability and risk modeling. Chapter 2 provides a general framework of the common reliability and risk models and their applications. Chapter 3 discusses reliability and risk models based on mixture distributions. Chapter 4 discusses general rules for reliability data analysis and also explores construction of reliability and risk models and estimation of the model parameters. Chapter 5 discusses load-strength (demand capacity) models and numerical methods for calculating the load-strength integral. Chapter 6 discusses Monte Carlo simulation algorithms for solving reliability and risk models, and Chapter 7 covers analysis of the properties of inhomogeneous media using Monte Carlo simulations. Chapter 8 discusses mechanisms of failure, including overstress failures, such as brittle fracture and ductile fracture; wear-out failures, such as fatigue and corrosion; and early-life failures, such as influence of the design. Chapter 9 discusses reliability associated with overstress failure mechanisms and damage factorization law. Chapter 10 introduces general equations related to the probability of failure of a stressed component with internal flaws and the individual probability of triggering failure with a single flaw. It also presents a stochastic model related to the fatigue life distribution of a component containing defects. Chapter 11 discusses the uncertainty and risk assessment associated with the location of the ductile-to-brittle transition region. Chapter 12 discusses modeling the kinetics of deterioration of protective coatings due to corrosion. Chapter 13 discusses physics-of-failure concepts that help improve the reliability of automotive suspension springs by delaying the fatigue failure mode. Chapter 14 discusses reliability governed by the relative locations of random variables in a finite domain. Chapter 15 discusses reliability based on minimum critical interval (MCI) and MFFOPs. It gives general equations related to random variables following a homogeneous Poisson process in a finite interval and some application examples. Chapter 16 provides a new methodology and models for reliability analysis and setting reliability requirements based on the cost of failure. Models and algorithms are introduced for determining the value from the reliability investment, the risk of premature failure, optimization models for minimizing the total losses, and models for limiting the risk of failure below a maximum acceptable level and for guaranteeing a minimum availability level. It is proved that the expected losses from failures of a repairable system in a specified time interval are equal to the expected number of failures times the expected cost given failure. Overall, this book examines the theory of reliability and risk analysis based on random variables. It also provides new methods for problem solving in the context of real reliability engineering problems. The book is ideal reading material for practicing engineers and consultants dealing with reliability and risk assessment. It also can be used as a reference by researchers and graduate students in reliability engineering and other quantitative disciplines such as actuarial science, economy, and applied probability and statistics.

Confidence in computational predictions is enhanced if the potential ‘error’ in these predictions (the difference between the prediction and nature’s outcome in the situation being simulated) can be credibly bounded. The “model-validation” process by which experimental or field results are compared to computational predictions to produce this confidence provides the raw material for characterizing a computational model’s predictive capability in terms of such error limits. In general, the goal is to evaluate predictive capability, first for predictions in the region of experimentation, then, if possible, for predictions in untested regions of applications. This whole process is fundamentally statistical because it requires the acquisition and careful analysis of appropriate data. We establish a statistical model for characterizing predictive-capability and discuss various experimental design and statistical data analysis issues and approaches for resolving them Analyses based on both ‘frequentist’ and Bayesian statistical paradigms are discussed in general in this paper and illustrated in accompanying papers presented at this workshop. The work of the first author was supported by Sandia National Laboratories and the United States Department of Energy under Contract DE-AC04-97AL85000. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockhed Martin Company, for the United States Department of Energy. **The work of the second author was supported by General Motors and the National Science Foundation, Grant DMS-0103265.

This report describes the underlying principles and goals of the Sandia ASCI Verification and Validation Program Validation Metrics Project. It also gives a technical description of two case studies, one in structural dynamics and the other in thermomechanics, that serve to focus the technical work of the project in Fiscal Year 2001.

Abstract The thermal conductivity of 304 stainless steel has been estimated from transient temperature measurements and knowing the volumetric heat capacity. Sensitivity coefficients were used to guide the design of this experiment as well as to estimate the confidence interval in the estimated thermal conductivity. The uncertainty on the temperature measurements was estimated by several means, and its impact on the estimated conductivity is discussed. The estimated thermal conductivity of 304 stainless steel is consistent with results from other sources.

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

The A-primed (Agile Product Realization of Innovative electro-Mechanical Devices) project is defining and proving processes for agile product realization for the Department of Energy complex. Like other agile production efforts reported in the literature, A-primed uses concurrent engineering and information automation technologies to enhance information transfer. A unique aspect of our approach to agility is the qualification during development of a family of related product designs and their production processes, rather than a single design and its attendant processes. Applying engineering principles and statistical design of experiments, economies of test and analytic effort are realized for the qualification of the device family as a whole. Thus the need is minimized for test and analysis to qualify future devices from this family, thereby further reducing the design-to-production cycle time. As a measure of the success of the A-primed approach, the first design took 24 days to produce, and operated correctly on the first attempt. A flow diagram for the qualification process is presented. Guidelines are given for implementation, based on the authors experiences as members of the A-primed qualification team.

Component-test plans are often designed by allocating system reliability among the system's components, then choosing individual component plans suitable for demonstrating achievement of each component's reliability goal. This approach does not consider how much information relative to the system reliability goal is provided by the ensemble of component tests. We consider the notion of system reliability operating characteristic (OC) curves, based on the component tests, and illustrate their use in designing or evaluating an overall test program. By specifying OC values (akin to producer's and consumer's risks), optimum, system-oriented component-test plans can be derived. These ideas are illustrated for a series system, and for a simple series-parallel system, with binomial data.

In a production process it is necessary to specify tolerances for characteristics, such as dimensions within which the measured characteristic must fall in order for the part to be acceptable. Such tolerances may often be set informally, but a consideration of measurement error and the attendant risks of incorrect decisions can be helpful in establishing appropriate tolerances. We consider five measures of consumer's risk and evaluate them with respect to computational convenience, information requirements, rationale, and economic considerations.

(1986). Discussion of Downing, Gardner, and Hoffman (1985) Technometrics: Vol. 28, No. 1, pp. 91-92.