Degree Type

Dissertation

Date of Award

1996

Degree Name

Doctor of Philosophy

Department

Industrial and Manufacturing Systems Engineering

First Advisor

Douglas W. McBeth

Abstract

This dissertation considers two developmental testing models for one-shot systems, non-repairable devices that are destroyed by testing or first normal use (for example, rocket engines) when there is the potential for reliability growth through redesign. Given a limited budget and fixed cost per redesign, we wish to determine the sequence of redesigns and tests so that the expected number of effective systems in an ultimate stockpile of systems of the final design is maximized. We develop mathematical models and analyses which describe the optimal testing policies for this sequential decision problem. Further, we present properties of the models which, in some cases, make the computation of solutions feasible. We begin with the analysis of a two-state reliability model. Besides determining the effect of redesign costs on the optimal strategy, we show how the possibility of harmful redesigns can be incorporated into the model. We explore how this model behaves when even a "poor" reliability design is highly reliable but extremely high reliability is desired. A primary contribution of the thesis is an analysis of the general multiple-state (k-state) reliability model and, most interestingly, the presentation of a 2-variable formulation by which some k-state models can be analyzed. Under some restrictions, we analyze and compute solutions to this latter model on a lattice.

DOI

https://doi.org/10.31274/rtd-180813-10374

Publisher

Digital Repository @ Iowa State University, http://lib.dr.iastate.edu/

Copyright Owner

Michael James Moon

Language

en

Proquest ID

AAI9626057

File Format

application/pdf

File Size

82 pages

Share

COinS