Date of Award

12-24-2015

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Department of Systems Engineering and Management

First Advisor

David R. Jacques, PhD.

Abstract

Organizations are often faced with portfolio construction efforts that require them to select one or more alternatives, subject to resource constraints, with the aim of achieving the maximum value possible. This is a well-defined problem with a number of analytically defensible approaches, provided the entire set of alternatives is known when the decision event takes place. Less well treated in the literature is how to approach this problem when the entire set of alternatives is unknown, as when the alternatives arrive over time. This change in the availability of data shifts the problem from one of identifying an optimal subset to one in which a series of smaller decisions are undertaken regarding the acceptability of each alternative as it presents itself. This work expands upon a methodology known as the Triage Method. The original Triage Method provided a screening tool that could be applied to alternatives as they presented themselves to determine if they should be accepted for further study, rejected out of hand, or held pending until later date. This decision was made strictly upon the value of the alternative and with no consideration of its cost. Two extensions to the Triage Method are offered which provide a capability to consider cost and other resource requirements of the alternatives, thus allowing a move from simply screening to portfolio selection. Guidelines are presented as to when each of these extensions is best employed, a characterization of the performance tradeoff between these and more traditional methodologies is developed, and insight and techniques for setting the value of parameters required by the extensions are provided.

AFIT Designator

AFIT-ENV-DS-15-D-018

DTIC Accession Number

AD1003566

Share

COinS