Abstract
Risk and reliability analyses of human errors or hardware failures sometimes need to enlist expert opinion in areas for which adequate human performance or hardware operational data are not available. Experts enlisted in this capacity provide probabilistic estimates of reliability, typically comprised of a measure of central tendency and uncertainty bounds. Formal guidelines for expert elicitation are readily available, but the methods are often time-consuming and costly to implement. This paper reports the first phase of a research effort to combine disparate formal methods of expert elicitation into a streamlined method. Fourteen reliability analysts were interviewed to identify current practices in expert elicitation.
Get full access to this article
View all access options for this article.