Many-attribute decision making using iterative attribute subsets
Ziegler, Laura E.
MetadataShow full item record
There are many approaches to multi-attribute decision making each with relative advantages and disadvantages, but some are either insufficient or not fully explored for many-attribute problems having an attribute set of more than two or three attributes. For instance, the Hypothetical Equivalents and Inequivalents Method (HEIM) can ensure consistent preferences and accommodate preferences from multiple stakeholders, but scales in complexity as the number of attributes increases. A large number of attributes makes it difficult for the decision maker to accurately process trade-offs, and the simultaneous scaling of both the dimension of the design space and the number of constraints results in issues related to visualization and efficiency. While previous work focused on a hierarchical representation of attributes in large environmental decision problems, the work presented in this thesis incorporates techniques that strategically elicit preferences over iterative subsets of attributes. Two case studies are presented to demonstrate how this process effectively adapts to the decision maker's preferences, constructing a systematic representation of preferences while successively eliminating less desirable design alternatives.