Suchergebnisse

Several recent nuclear power plant probabilistic risk assessments (PRAs) have utilized broadened Reactor Safety Study (RSS) component failure rate population variability curves to compensate for such things as expert "overvaluation bias" in the estimates upon which the curves are based.A simple two‐components of variation empirical Bayes model is proposed for use in estimating the between‐expert variability curve in the presence of such biases. Under certain conditions this curve is a population variability curve. Comparisons are made with the existing method.The popular procedure appears to be generally much more conservative than the empirical Bayes method in removing such biases. In one case the broadened curve based on the popular method is more than two orders of magnitude broader than the empirical Bayes curve. In another case it is found that the maximum justifiable degree of broadening of the RSS curve is to increase α from 5% to 12%, which is significantly less than the 20% value recommended in the popular approach.

AbstractEmpirical Bayes single‐sample acceptance sampling procedures are derived. These procedures assume that the lot fraction defective varies randomly according to a completely unknown and unspecified prior distribution. The unknown prior density function is estimated based on sampling results from previous lots. A procedure is developed for obtaining single‐sampling plans that achieve specified posterior consumer and producer risks. The procedures are illustrated for a real‐data example.

A quantitative probabilistic/systems analysis model is described which is useful for allocating resources to safeguard valuable documents or materials in either a fixed‐site facility or a moving convoy against an overt terrorist attack. The model is also useful for ranking the sensitive areas at a site according to their survivability of a given hypothesized terrorist attempt. To compare various defense strategies and security configurations, the probability of a successful terrorist activity is computed based on event tree models of the site/security configuration. This calculation incorporates a realistic engagement model (in the event a guard force engages the terrorists prior to completion of their objective) and information on barrier penetration times (for example, distribution of the time to defeat a chain link fence or vault door, traverse an open area, and so forth). Two security analyses are described to illustrate the methodology. One example considers a terrorist attack on a convoy transporting a missile from a storage to a launch facility. The second example involves an attack on a munitions storage facility.