Advantages of variance reduction techniques in establishing confidence intervals for quantiles

Abstract

Over the past two decades, U.S. nuclear power plant regulation has increasingly depended on best-estimate plus uncertainty safety analyses. As a result of the shift to best-estimate analyses, the distribution of the output metric must be compared against a regulatory goal, rather than a single, conservative value. This comparison has historically been conducted using a 95% one-sided confidence interval for the 0.95-quantile of the output distribution, which is usually found following the technique of simple random sampling using order statistics (SRS-OS). While SRS-OS has certain statistical advantages, there are drawbacks related to the available sampling schemes and the accuracy and precision of the resulting value. Recent work has shown that it is possible to establish asymptotically valid confidence intervals for a quantile of the output of a model simulated using variance reduction techniques (VRTs). These VRTs can provide more informative results than SRS-OS. This work compares SRS-OS and the VRTs of antithetic variates and Latin hypercube sampling through several experiments, designed to replicate conditions found in nuclear safety analyses. This work is designed as an initial investigation into the use of VRTs as a tool to satisfy nuclear regulatory requirements, with hope of expanded analyses of VRTs in the future.

Reliability Evaluation and Fault Tolerant Design for KLL Sketches

“…This process is called compaction. In particular, when the H layer is compacted, the top layer is set as H = H+1, and the array is expanded according to equation (1). When the first layer is extracted, it should be sorted first.…”

“…uantile estimation is of great importance in many application domains such as engineering systems, risk analysis for social, environmental, and financial systems or big data analytics [1]. For example, in datacenters, the 0.5 and 0.95 quantiles are often used to monitor the typical and abnormal behavior of the system [2].…”

Reliability modeling of subsea SISs partial testing subject to delayed restoration

“…More details about Petri nets can be found in [4,5,11,39]. The Petri net module in the GRIF software [38] is employed to model the behavior of complex dynamic systems for performance evaluation with 95% confidence intervals, and they are calculated to provide a more practical explanation as well as to better assess the failure distribution of the system [40,41]. 5.1 Petri net modeling for 1oo1 and 1oo2 systems RBD driven Petri net models for 1oo1 system and 1oo2 system [29] are developed under different testing strategies considering MRT as presented in Fig.…”

Uncertainty analysis of a large break loss of coolant accident in a pressurized water reactor using non-parametric methods

“…These FOMs are usually extreme values (minima, maxima) of safety variables during the transient, such as Peak Clad Temperature (PCT), Critical Heat Flux (CHF), etc. Current BEPU methodologies mainly rely on a probabilistic description of the uncertainty and on the use of statistical techniques to estimate it [9,10,11,12,13]. In this framework, the uncertainty of a FOM can be identified with its probability distribution.…”