skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Elements of Bayesian experimental design

Abstract

We consider some elements of the Bayesian approach that are important for optimal experimental design. While the underlying principles used are very general, and are explained in detail in a recent tutorial text, they are applied here to the specific case of characterising the inferential value of different resolution peakshapes. This particular issue was considered earlier by Silver, Sivia and Pynn (1989, 1990a, 1990b), and the following presentation confirms and extends the conclusions of their analysis.

Authors:
 [1]
  1. Rutherford Appleton Lab., Oxon (United Kingdom)
Publication Date:
Research Org.:
Lawrence Berkeley National Lab., CA (United States); Los Alamos National Lab., NM (United States)
OSTI Identifier:
569579
Report Number(s):
LBNL-40816; CONF-9609353-
ON: DE98050032; TRN: 98:003949
Resource Type:
Conference
Resource Relation:
Conference: Workshop on methods for neutron scattering instrumentation design, Berkeley, CA (United States), 23-25 Sep 1996; Other Information: PBD: Sep 1997; Related Information: Is Part Of Proceedings of a workshop on methods for neutron scattering instrumentation design; Hjelm, R.P. [ed.] [Los Alamos National Lab., NM (United States)]; PB: 150 p.
Country of Publication:
United States
Language:
English
Subject:
44 INSTRUMENTATION, INCLUDING NUCLEAR AND PARTICLE DETECTORS; COMPUTER-AIDED DESIGN; NEUTRON DIFFRACTOMETERS; NEUTRON SPECTROMETERS; COMPUTERIZED SIMULATION

Citation Formats

Sivia, D.S. Elements of Bayesian experimental design. United States: N. p., 1997. Web.
Sivia, D.S. Elements of Bayesian experimental design. United States.
Sivia, D.S. 1997. "Elements of Bayesian experimental design". United States. doi:. https://www.osti.gov/servlets/purl/569579.
@article{osti_569579,
title = {Elements of Bayesian experimental design},
author = {Sivia, D.S.},
abstractNote = {We consider some elements of the Bayesian approach that are important for optimal experimental design. While the underlying principles used are very general, and are explained in detail in a recent tutorial text, they are applied here to the specific case of characterising the inferential value of different resolution peakshapes. This particular issue was considered earlier by Silver, Sivia and Pynn (1989, 1990a, 1990b), and the following presentation confirms and extends the conclusions of their analysis.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 1997,
month = 9
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Abstract not provided.
  • Abstract not provided.
  • In an analysis of probabilistic risk, safety, and reliability of a nuclear power plant, the reliability data base (DB) must be established first. As the importance of the reliability data base increases, event reporting systems such as the US Nuclear Regulatory Commission's Licensee Event Report and the International Atomic Energy Agency's Incident Reporting System have been developed. In Korea, however, the systematic reliability data base is not yet available. Therefore, foreign data bases have been directly quoted in reliability analyses of Korean plants. In order to develop a reliability data base for Korean plants, the problem is which methodology ismore » to be used, and the application limits of the selected method must be solved and clarified. After starting the commercial operation of Korea Nuclear Unit-1 (KNU-1) in 1978, six nuclear power plants have begun operation. Of these, only KNU-3 is a Canada Deuterium Uranium pressurized heavy-water reactor, and the others are all pressurized water reactors. This paper describes the proposed reliability data-base network (KNRDS) for Korean nuclear power plants in the context of two-stage Bayesian (TSB) procedure of Kaplan. It describes the concept of TSB to obtain the Korean-specific plant reliability data base, which is updated with the incorporation of both the reported generic reliability data and the operation experiences of similar plants.« less
  • In a computational experiment, the data are produced by a computer program that models a physical system. The experiment consists of a set of model runs; the design of the experiment specifies the choice of program inputs for each run. This paper centers on the problem of prediction (interpolation), the goal of which is to devise a design/analysis method which will provide predictions of model output for input values not run. We adopt a Bayesian approach as the basis for the analysis. Uncertainty about the response function is quantified by choosing a class of probability distributions over the function space.more » This leads to design procedures based on maximizing the expected reduction in ''amount of uncertainty,'' where the latter can be defined formally in terms of properties of the posterior distribution. Here we use as a design optimality criterion the determinant of the posterior covariance matrix of the responses at the input configurations at which we want to make predictions. This requires maximization of the determinant of the prior covariance matrix of the responses at the design sites. We describe our computer algorithm for constructing optimal designs, and give some examples of designs that it produces. 9 refs., 1 fig., 1 tab.« less