1. Introduction to Bayesian Methods in Reliability.- 1. Why Bayesian Methods?.- 1.1 Sparse data.- 1.2 Decision problems.- 2. Bayes' Theorem.- 3. Examples from a Safety Study on Gas transmission Pipelines.- 3.1 Estimating the probability of the development of a big hole.- 3.2 Estimating the leak rate of a gas transmission pipeline.- 4. Conclusions.- References.- 2. An Overview of the Bayesian Approach.- 1. Background.- 2. Probability Concepts.- 3. Notation.- 4. Reliability Concepts and Models.- 5. Forms of Data.- 6. Statistical Problems.- 7. Review of Non-Bayesian Statistical Methods.- 8. Desiderata for Decision-Oriented Statistical Methodology.- 9. Decision-Making.- 10. Degrees of Belief as Probabilities.- 11. Bayesian Statistical Philosophy.- 12. A Simple Illustration of Bayesian Learning.- 13. Bayesian Approaches to Typical Statistical Questions.- 14. Assessment of Prior Densities.- 15. Bayesian Inference for some Univariate Probability Models.- 16. Approximate Analysis under Great Prior Uncertainty.- 17. Problems Involving many Parameters: Empirical Bayes.- 18. Numerical Methods for Practical Bayesian Statistics.- References.- 3. Reliability Modelling and Estimation.- 1. Non-Repairable Systems.- 1.1 Introduction.- 1.2 Describing reliability.- 1.3 Failure time distributions.- 2. Estimation.- 2.1 Introduction.- 2.2 Classical methods.- 2.3 Bayesian methods.- 3. Reliability estimation.- 3.1 Introduction.- 3.2 Binomial sampling.- 3.3 Pascal sampling.- 3.4 Poisson sampling.- 3.5 Hazard rate estimation.- References.- 4. Repairable Systems and Growth Models.- 1. Introduction.- 2. Good as New: the Renewal Process.- 3. Estimation.- 4. The Poisson Process.- 5. Bad as old: the Non-Homogeneous Poisson Process.- 6. Classical Estimation.- 7. Exploratory Analysis.- 8. The Duane Model.- 9. Bayesian Analysis.- References.- 5. The Use of Expert Judgement in Risk Assessment.- 1. Introduction.- 2. Independence Preservation.- 3. The Quality of Experts' Judgement.- 4. Calibration Sets and Seed Variables.- 5. A Classical Model.- 6. Bayesian Models.- 7. Some Experimental Results.- References.- 6. Forecasting Software Reliability.- 1. Introduction.- 2. The Software Reliability Growth Problem.- 3. Some Software Reliability Growth Models.- 3.1 Jelinski and Moranda (JM).- 3.2 Bayesian Jelinski-Moranda (BJM).- 3.3 Littlewood (L).- 3.4 Littlewood and Verrall (LV).- 3.5 Keiller and Littlewood (KL).- 3.6 Weibull order statistics (W).- 3.7 Duane (D).- 3.8 Goel-Okumoto (GO).- 3.9 Littlewood NHPP (LNHPP).- 4. Examples of Use.- 5. Analysis of Predictive Quality.- 5.1 The u-plot.- 5.2 The y-plot, and scatter plot of u's.- 5.3 Measures of 'noise'.- 5.3.1 Braun statistic.- 5.3.2 Median variability.- 5.3.3 Rate variability.- 5.4 Prequential likelihood.- 6. Examples of Predictive Analysis.- 7. Adapting and Combining Predictions; Future Directions.- 8 Summary and Conclusions.- Acknowledgements.- References.- References.- Author index.

When data is collected on failure or survival a list of times is obtained. Some of the times are failure times and others are the times at which the subject left the experiment. These times both give information about the performance of the system. The two types will be referred to as failure and censoring times (cf. Smith section 5). * A censoring time, t, gives less information than a failure time, for it is * known only that the item survived past t and not when it failed. The data is tn and of censoring thus collected as a list of failure times t , . . . , l * * * times t , t , . . . , t * 1 z m 2. 2. Classical methods The failure times are assumed to follow a parametric distribution F(t;B) with and reliability R(t;B). There are several methods of estimating density f(t;B) the parameter B based only on the data in the sample without any prior assumptions about B. The availability of powerful computers and software packages has made the method of maximum likelihood the most popular. Descriptions of most methods can be found in the book by Mann, Schafer and Singpurwalla (1974). In general the method of maximum likelihood is the most useful of the classical approaches. The likelihood approach is based on constructing the joint probability distrilmtion or density for a sample.