"Consumer preference for any product or product feature can be expressed in the form of a utility function. Many such utility functions form a part of a preference map, where each of these are expressed in terms of the attributes defining the product or the product feature. In order to optimize the design, it is required to optimize the overall utility function obtained by a mathematical combination of individual utility functions defined in the preference map. The objective of this research is to devise and implement an algorithm to optimize all the individual utility functions comprised in a preference map for a product or product feature. Executed together, this will optimize the overall utility function, U(x). So, an algorithm is needed to compute the optimal values for each attribute forming the individual utility functions by efficiently and thoroughly testing the entire allowed range of values in the function domain, i.e. the global optimum. The challenges faced in this include the presence of a complex space created by interactions between the various attributes in the preference map. This makes it prohibitive to solve using traditional algorithms. Thus, software agents aid in the computation as two or more software agents can collaborate on the task of optimization, enabling every single software agent to cater to a single attribute. Thus, any number of software agents can be employed to run synchronously so that all the concerned attributes can be efficiently optimized"--Abstract, page iii.
Liu, Xiaoqing Frank
Orsborn, Seth D.
Madria, Sanjay Kumar
M.S. in Computer Science
Missouri University of Science and Technology
viii, 46 pages
© 2010 Pavitra Dhruvanarayana, All rights reserved.
Thesis - Open Access
Library of Congress Subject Headings
Ants -- Behavior -- Mathematical models
Consumers' preferences -- Mathematical models
Print OCLC #
Electronic OCLC #
Link to Catalog Recordhttp://laurel.lso.missouri.edu/record=b8456940~S5
Dhruvanarayana, Pavitra, "Dynamic ant colony optimization for globally optimizing consumer preferences" (2010). Masters Theses. 5126.