Reliability: Modeling, Prediction, and Optimization

CHAPTER 3

Collection and Preliminary Analysis of Failure Data

3.1 INTRODUCTION

In Chapter 1, the two concepts most important to this book, failure and reliability, were introduced. There are many aspects to failure—time to failure, failure mode, causes of failure, consequences and their severity, and so forth—and many dimensions of reliability—theory, modeling, test design, assessment, optimization, and so forth. In this context, the acquisition, analysis, and interpretation of data play a key role in wedding theory and practice, mathematics and the real world.

It is apparent from the examples in Chapter 1 and the cases in Chapter 2 that many types of data are encountered in reliability applications and that the range of applications is quite extensive, including electronics, medicine, aerospace, automotive, heavy industry, simple consumer products, power tools, appliances, and so forth, i.e., nearly every industrial and commercial enterprise. One of the objectives of this chapter is to categorize these data types and indicate other possible sources of information that may be useful to engineers, analysts, and managers in the analysis of reliability. These range from test data to operational data to subjective data (“engineering judgment”).

In most cases, we will be dealing with data on time to failure. Such data are variously called “failure data,” “lifetime data,” “life data,” and so forth. In more general applications (and sometimes in reliability), they may simply be time until ...