Dissertation/Thesis Abstract

Use of probabilistic inversion to model qualitative expert input when selecting a new nuclear reactor technology
by Merritt, Charles R., Jr., D.Sc., The George Washington University, 2008, 112; 3297448
Abstract (Summary)

Complex investment decisions by corporate executives often require the comparison of dissimilar attributes and competing technologies. A technique to evaluate qualitative input from experts using a Multi-Criteria Decision Method (MCDM) is described to select a new reactor technology for a merchant nuclear generator. The high capital cost, risks from design, licensing and construction, reactor safety and security considerations are some of the diverse considerations when choosing a reactor design. Three next generation reactor technologies are examined: the Advanced Pressurized-1000 (AP-1000) from Westinghouse, Economic Simplified Boiling Water Reactor (ESBWR) from General Electric, and the U.S. Evolutionary Power Reactor (U.S. EPR) from AREVA. Recent developments in MCDM and decision support systems are described. The uncertainty inherent in experts' opinions for the attribute weighting in the MCDM is modeled through the use of probabilistic inversion. In probabilistic inversion, a function is inverted into a random variable within a defined range. Once the distribution is created, random samples based on the distribution are used to perform a sensitivity analysis on the decision results to verify the "strength" of the results. The decision results for the pool of experts identified the U.S. EPR as the optimal choice.

Indexing (document details)
Advisor: Mazzuchi, Thomas A., Sarkani, Shahram
Commitee: Carr, Matthew A., Murphree, Edward L., Jr., Ryan, Julie J.
School: The George Washington University
Department: Engineering Mgt and Systems Engineering
School Location: United States -- District of Columbia
Source: DAI-B 69/02, Dissertation Abstracts International
Subjects: Industrial engineering, Energy
Keywords: Expert judgment, Multicriteria decision-making, Nuclear power, Probabilistic inversion, Sensitivity analysis
Publication Number: 3297448
ISBN: 978-0-549-46068-8
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy