Michigan State University
Paletteviz : a method for visualization of high-dimensional pareto-optimal front and its applications to multi-criteria decision making and analysis
Abstract
dc:descriptionVisual representation of a many-objective Pareto-optimal front in four or more dimensional objective space requires a large number of data points. Moreover, choosing a single point from a large set even with certain preference information is problematic, as it imposes a large cognitive burden on the decision-makers. Therefore, many-objective optimization and decision-making practitioners have been interested in effective visualization methods to en- able them to filter down a large set to a few critical points for further analysis. Most existing visualization methods are borrowed from other data analytics domains and they are too generic to be effective for many-criterion decision making. In this dissertation, we propose a visualization method, using star-coordinate and radial visualization plots, for effectively visualizing many-objective trade-off solutions. The proposed method respects some basic topological, geometric and functional decision-making properties of high-dimensional trade- off points mapped to a three-dimensional space. We call this method Palette Visualization (PaletteViz). We demonstrate the use of PaletteViz on a number of large-dimensional multi- objective optimization test problems and three real-world multi-objective problems, where one of them has 10 objective and 16 constraint functions. We also show the uses of NIM-BUS and Pareto-Race concepts from canonical multi-criterion decision making and analysis literature and introduce them into PaletteViz to demonstrate the ease and advantage of the proposed method.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Talukder, Akm. Khaled Ahsan
- Contributors dc:contributor
-
- Deb, Kalyanmoy
- Goodman, Erik
- Kulkarni, Sandeep
- Zhou, Jiayu
Subjects
dc:subject × 3Rights
dc:rights- Statement dc:rights
-
- Attribution 4.0 International
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
etd:50500
isbn:9798837532139
oclc:on1373601405
oclc:1373601405
umi:28970536
local:Talukder_grad.msu_0128D_18899